Once satisfied with annotated results, you can proceed with running the full pipeline. -***NOTE:*** If you modified any methods, be sure to update the predefined code to use your changes in the pipeline. +***NOTE:*** If you modified any methods within the test script, be sure to update the predefined code to use your changes in the pipeline.
@@ -200,48 +187,109 @@ Please note the model labels are retrieved from the model directly, so the model ## Testing Pipeline Components We provided multiple test scripts to test different components of the pipeline. -Each of the tests can be run as specified in the [Smart Filtering Pipeline: Testing](#testing) section using the same video `anduril_swarm_8K.mp4` as `INPUT_STREAM`. +Each of the tests can be run as specified in the [Smart Filtering Pipeline: Testing](#testing) section using the same video `anduril_swarm_8K.mp4` as `SOURCE`. Here we provide details on each available test. -| Test File | Description | -| --------- | ----------- | -| test_readers.py | Test the readers for both CPU and GPU on provided RTSP URL or video file | -| test_model.py | Test the model for both CPU and GPU on provided RTSP URL or video file | - - -### Test Readers -This test reads the `INPUT_STREAM` from `./inputs`, `/watch_dir`, or RTSP server, and uses the readers to process frames for length of video or 2 minutes (whichever comes first). -```bash -python test_readers.py --source "${INPUT_STREAM}" --duration 2 -``` +| Component | Test File | Description | +| --------- | --------- | ----------- | +| Detection Model | [test_model.py](/fastapi/tests/test_model.py) | Test the model for both CPU and GPU on provided RTSP URL or video file | +| Smart Filtering | [test_detections.py](/fastapi/tests/test_detections.py) | This test independently evaluates the detection pipeline (with and without Smart Filtering) only and does not include video clip generation or sending metadata to database for querying. | +| Stream Reader | [test_pipeline.py](/fastapi/tests/test_pipeline.py) | Scenario 1 tests the behavior of Readers when provided an invalid RTSP url.
Scenario 2 reads the RTSP url or video file for a specified duration or until it ends. | +| Video Clip Generation | [test_pipeline.py](/fastapi/tests/test_pipeline.py) | Scenario 3 mimics the clip generation within the pipeline. | +| Smart Filtering | [test_pipeline.py](/fastapi/tests/test_pipeline.py) | Scenario 4 tests the entire pipeline and saves output video of results. | ### Test Models -This test reads the `INPUT_STREAM` from `./inputs`, `/watch_dir`, or RTSP server, and creates a video with overlaid drone detection results for both GPU and CPU model. +This test reads the `SOURCE` from `./inputs` (the local directory or `/watch_dir` if in container), or RTSP server, and creates a video with overlaid detection results for both GPU and CPU model. ```bash -python test_models.py -v "${INPUT_STREAM}" +python test_models.py -s "${SOURCE}" ``` You can also use the Yolo11n model by using: ```bash -python test_models.py -v "${INPUT_STREAM}" -m yolo11n --no-custom +python test_models.py -s "${SOURCE}" -m yolo11n --no-custom ``` +The arguments available for this test are as follows: +| Argument | Default | Description | +| -------- | ------- | ----------- | +| -s SOURCE,
--source SOURCE | anduril_swarm_8K.mp4 | Video filename (located in /inputs) or RTSP target stream endpoint | +| --no-custom | - | Enable if using Ultralytics YOLO model | +| -m MODEL_NAME,
--model MODEL_NAME | drone_detection | Name of model. Required if `--no-custom` is enabled. | +| --type {object,motion} | None | Filter by detection type (object or motion) | +| --device {cpu,gpu} | None | Filter by device (cpu or gpu) | +
+ -### Test Video Clipping -This test reads the `INPUT_STREAM` from `./inputs`, or `/watch_dir` and creates 10 second video clips. +### Test Detections +This test reads the `SOURCE` from `./inputs`, and evaluates the detection pipeline only. + +To run the detection test, use the following: ```bash -python test_clipping.py -v "${INPUT_STREAM}" +python test_detections.py --source "${SOURCE}" ``` +You also have control on which scenario, model, etc. to use during testinh. +The arguments available are as follows: +| Argument | Default | Description | +| -------- | ------- | ----------- | +| -s SOURCE,
--source SOURCE | anduril_swarm_8K.mp4 | Video filename (located in ./inputs) | +| --no-custom | - | Enable if using Ultralytics YOLO model | +| -m MODEL_NAME,
--model MODEL_NAME | drone_detection | Name of model. Required if `--no-custom` is enabled. | +| --type {object,motion} | None | Filter by detection type (object or motion). "motion" shows the ROI while "object" shows detection results. | +| --device {cpu,gpu,all} | all | Filter by target hardware. | +| --sf | None | Filter test by Smart Filtering pipeline | +| --debug | False | Enable debug message | +| -n DEBUG_FRAME_LIMIT | 100 | Number of frames used for debugging | +
+ -### Test Smart Filtering -This is the test mentioned above in [Smart FIltering Pipeline: Testing](#testing). -The test reads the `INPUT_STREAM` from `./inputs`, or `/watch_dir` and applies the Smart Filtering pipeline. +### Test Pipeline +This test reads the `SOURCE` from `./inputs` or RTSP server, and uses the stream readers to perform multiple scenarios. + +To run the full pipeline test, use the following: ```bash -python test_detections.py -v "${INPUT_STREAM}" +python test_pipeline.py --source "${SOURCE}" ``` +You also have control on which scenario, model, etc. to use during testinh. +The arguments available are as follows: +| Argument | Default | Description | +| -------- | ------- | ----------- | +| -s SOURCE,
--source SOURCE | anduril_swarm_8K.mp4 | Video filename (located in ./inputs) or RTSP target stream endpoint | +| -d DURATION,
--duration DURATION | 2 | Test duration in minutes | +| --scenario | - | Specify one or more scenarios. Otherwise all scenarios are ran. Available scenarios: [1, 2, 3, 4] | +| --no-custom | - | Enable if using Ultralytics YOLO model | +| -m MODEL_NAME,
--model MODEL_NAME | drone_detection | Name of model. Required if `--no-custom` is enabled. | +| --type {object,motion} | None | Filter by detection type (object or motion). "motion" shows the ROI while "object" shows detection results. | +| --device {cpu,gpu,all} | all | Filter by target hardware. | +| --sf | None | Filter test by Smart Filtering pipeline | +| --debug | False | Enable debug message | + +
+ +#### Scenario 1: Invalid RTSP URL +Scenario 1 is a simple test to verify the pipeline ends if an invalid input is provided. For this test, an invalie RTSP url is provided, there are 5 retry attempts, and after a failed connection, the test ends. By default, the test iterates over available devices ("cpu" and "gpu"). + + +#### Scenario 2: Stability & Throughput +Scenario 2 evaluates the stability and throughput of the readers by continuously reading from the reader for the duration of the source or for a specified duration (whichever comes first). By default, the test iterates over available devices ("cpu" and "gpu"). + + +#### Scenario 3: Video Clip Generation +Scenario 3 evaluates the video clip generation process in the pipeline. +In the pipeline, we segment the input video into 10 sec clips for easy retrieval and playback associated with results in the query UI. By default, the test iterates over available devices ("cpu" and "gpu"). +The resulting video clips generated from this test are located in `test_pipeline_results/scenario3_*`. + + +#### Scenario 4: Pipeline (Without Sending Metadata) +Scenario 4 mimics the entire pipeline excluding sending the generated metadata to VDMS. By default, the test iterates over available devices ("cpu" and "gpu"), detection type ("object" or "motion"), and with and without SF. +The resulting video clips and the detection results are located in `test_pipeline_results/scenario4_*`. + +In the case of testing the Smart Filtering pipeline, to display the ROIs, you can use the following, which will test for both gpu and cpu. +```bash +python test_pipeline.py --scenario 4 --source "${SOURCE}" --type motion +``` ## Pipeline Deployment diff --git a/fastapi/.vscode/launch.json b/fastapi/.vscode/launch.json new file mode 100644 index 0000000..5ea2550 --- /dev/null +++ b/fastapi/.vscode/launch.json @@ -0,0 +1,79 @@ +{ + // Use IntelliSense to learn about possible attributes. + // Hover to view descriptions of existing attributes. + // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387 + "version": "0.2.0", + "configurations": [ + { + "name": "Python Debugger: Pipeline", + "type": "debugpy", + "request": "launch", + "program": "${workspaceFolder}/tests/test_pipeline.py", + "console": "integratedTerminal", + "subProcess": false, + "justMyCode": false, + "args": [ + "--duration", "2", + + // AVAILABLE SCENARIO PARAMS + // Use --scenario to specify tests by number (1 - 4) + // 1 - 3: --device (cpu or gpu) + // 4: --device, --sf, --type (object or motion) + + // VIDEOS EXPECTED TO BE rstp url OR IN inputs/ or /watch_dir + // "--source", "rtsp://
Live Detection Dashboard
+ + + + +
+
+
+
+
+
+ {% for id in cameras %}
+
+
+
+
+
\ No newline at end of file
diff --git a/fastapi/tests/nginx.conf b/fastapi/tests/nginx.conf
new file mode 100644
index 0000000..ac1b556
--- /dev/null
+++ b/fastapi/tests/nginx.conf
@@ -0,0 +1,12 @@
+events { worker_connections 1024; }
+
+http {
+ include mime.types;
+ default_type application/octet-stream;
+ client_max_body_size 100M;
+
+ server {
+ listen 80;
+ server_name _;
+ }
+}
\ No newline at end of file
diff --git a/fastapi/tests/test_detections.py b/fastapi/tests/test_detections.py
new file mode 100644
index 0000000..6cac5b9
--- /dev/null
+++ b/fastapi/tests/test_detections.py
@@ -0,0 +1,2179 @@
+import argparse
+import csv
+import gc
+import logging
+import multiprocessing as mp
+import os
+import sys
+import threading
+import time
+from pathlib import Path
+
+import cv2
+import matplotlib.pyplot as plt
+import numpy as np
+import pytest
+import tensorrt as trt
+import torch
+import torch.nn.functional as F
+
+sys.path.insert(1, str(Path(__file__).parent.parent))
+from include.default_configs import (
+ CUSTOM_MODEL_FLAG_DEFAULT,
+ DEBUG_DEFAULT,
+ MODEL_NAME_DEFAULT,
+ THRESHOLD_VALUE,
+)
+from include.handlers import (
+ CPUStreamHandler,
+ DeviceBaseHandler,
+ GPUStreamHandler,
+ log_to_logger,
+)
+from include.handlers import test_rendering_worker as rendering_worker
+from include.utils import (
+ PipelineConfig,
+ tensor2opencv,
+)
+
+try:
+ torch.multiprocessing.set_start_method("spawn", force=True)
+except RuntimeError:
+ pass
+
+logging.getLogger("matplotlib").setLevel(logging.WARNING)
+logger = trt.Logger(trt.Logger.WARNING)
+trt.init_libnvinfer_plugins(logger, "")
+
+DEBUG_FLAG_DEFAULT = True if DEBUG_DEFAULT == "1" else False
+os.environ["OMP_NUM_THREADS"] = "1"
+os.environ["PYTORCH_ALLOC_CONF"] = "expandable_segments:True"
+force_export = False
+
+
+# def render_display_worker(queue, output_path, fps, size):
+# """Background process that draws labels and writes to video."""
+# fourcc = cv2.VideoWriter_fourcc(*"avc1")
+# writer = cv2.VideoWriter(output_path, fourcc, fps, size)
+
+# if not writer.isOpened():
+# print(f" [ERROR] VideoWriter failed to open: {output_path}")
+# return
+
+# while True:
+# item = queue.get()
+# if item is None:
+# break
+
+# display_frame, metadata_or_bbs, class_list = item
+# if (display_frame.shape[1], display_frame.shape[0]) != size:
+# display_frame = cv2.resize(display_frame, size)
+
+# if isinstance(metadata_or_bbs, dict):
+# for _, obj in metadata_or_bbs.items():
+# bbox = obj["bbox"]
+# x, y, w, h = bbox["x"], bbox["y"], bbox["width"], bbox["height"]
+# class_name = bbox["object"]
+# class_id = class_list.index(class_name) if class_name in class_list else 0
+# confidence = bbox["object_det"]["confidence"]
+
+# bb_color = get_detection_color(class_id, is_bgr=True)
+# label = f"{class_name} {confidence:.2f}"
+
+# cv2.rectangle(display_frame, (x, y), (x + w, y + h), bb_color, 2)
+# draw_label(display_frame, label, (x, y), color=bb_color, padding=5)
+# elif metadata_or_bbs is not None:
+# for box in metadata_or_bbs:
+# if torch.is_tensor(box):
+# box = box.tolist()
+# x1, y1, x2, y2 = map(int, box)
+# cv2.rectangle(display_frame, (x1, y1), (x2, y2), (0, 0, 225), 2)
+
+# writer.write(display_frame)
+
+# writer.release()
+
+
+def fps_comparison_chart(chart_path, results, fps_key="Pipeline FPS (Video frames)"):
+ try:
+ names = [r["Test Name"] for r in results]
+ fps_values = [float(r[fps_key]) for r in results]
+
+ plt.figure(figsize=(10, 6))
+ plt.grid(axis="y", linestyle="--", alpha=0.7, zorder=0)
+
+ colors = ["#2ca02c" if "gpu" in n.lower() else "#1f77b4" for n in names]
+ bars = plt.bar(names, fps_values, color=colors, zorder=3)
+ plt.ylabel("Frames Per Second (FPS)")
+ plt.title(f"Performance Comparison: {chart_path.stem}")
+ plt.xticks(rotation=45)
+
+ for bar in bars:
+ yval = bar.get_height()
+ plt.text(
+ bar.get_x() + bar.get_width() / 2,
+ yval + 1,
+ f"{yval:.1f}",
+ ha="center",
+ va="bottom",
+ )
+
+ if fps_values:
+ plt.ylim(0, max(fps_values) * 1.2)
+
+ plt.tight_layout()
+ plt.savefig(str(chart_path))
+ print(f" Comparison chart saved to: {chart_path}")
+ except Exception:
+ print("Skipping chart generation: error occurred.")
+
+
+@pytest.fixture(scope="class")
+def setup_context(request):
+ """Replaces setUpClass: Runs once per test class."""
+ current_test_filename = Path(__file__).stem
+ test_dir = Path(__file__).parent
+ main_path = test_dir.parent
+ video_dir = main_path / "inputs"
+
+ VIDEO_FILENAME = os.getenv("VIDEO_FILENAME", "anduril_swarm_8K.mp4")
+ if video_dir.exists():
+ request.cls.video_path = video_dir / VIDEO_FILENAME
+ else:
+ video_dir = Path("/watch_dir")
+ request.cls.video_path = video_dir / VIDEO_FILENAME
+
+ model_name = os.getenv("MODEL_NAME", MODEL_NAME_DEFAULT)
+ request.cls.result_dir = (
+ test_dir
+ / f"{current_test_filename}_results/{model_name}"
+ / request.cls.video_path.stem
+ )
+ request.cls.result_dir.mkdir(parents=True, exist_ok=True)
+ request.cls.benchmarks = []
+ request.cls.csv_filename = (
+ f"pipeline_benchmarks_{model_name}_{request.cls.video_path.stem}.csv"
+ )
+
+ request.cls.name = request.cls.video_path.stem
+ request.cls.source = str(request.cls.video_path)
+ request.cls.active = True
+ request.cls.active_streams = {}
+
+ request.cls._shared_model = None
+ request.cls._shared_model_path = None
+ request.cls._shared_model_device = None
+ request.cls._shared_model_sf_enabled = None
+ yield
+
+ if request.cls.benchmarks:
+ # Filter and exclude rows that were interrupted or failed initialization due to an early pytest skip
+ request.cls.benchmarks = [
+ r for r in request.cls.benchmarks if r and "Test Name" in r
+ ]
+ results = request.cls.benchmarks
+ for row in results:
+ if "gpu" in row["Test Name"].lower():
+ match_name = row["Test Name"].replace("gpu", "cpu")
+ cpu_row = next(
+ (r for r in results if r["Test Name"] == match_name), None
+ )
+ if cpu_row:
+ gpu_fps = float(row["Pipeline FPS (Video frames)"])
+ cpu_fps = float(cpu_row["Pipeline FPS (Video frames)"])
+ speedup = (gpu_fps / cpu_fps) if cpu_fps > 0 else 0
+ row["Pipeline Speedup vs CPU"] = f"{speedup:.2f}x"
+ else:
+ row["Pipeline Speedup vs CPU"] = "N/A"
+ else:
+ row["Pipeline Speedup vs CPU"] = "Baseline (CPU)"
+
+ keys = results[0].keys()
+ with open(
+ str(request.cls.result_dir / request.cls.csv_filename), "w", newline=""
+ ) as f:
+ dict_writer = csv.DictWriter(f, fieldnames=keys)
+ dict_writer.writeheader()
+ dict_writer.writerows(results)
+
+ print(f"\n[FINAL] Benchmarks saved to {request.cls.csv_filename}")
+
+ for r in results:
+ print(
+ f" > {r['Test Name']}: {r['Pipeline FPS (Video frames)']} FPS | {r['Pipeline FPS (Target frames)']} FPS | {r['sf+roi+det FPS (Target frames)']} FPS | {r['Display FPS']} FPS | Speedup: {r.get('Pipeline Speedup vs CPU', 'N/A')}"
+ )
+
+ chart_path = (
+ request.cls.result_dir
+ / f"{request.cls.csv_filename.replace('.csv', '')}_sfFPS.png"
+ )
+ fps_comparison_chart(
+ chart_path, results, fps_key="sf+roi+det FPS (Video frames)"
+ )
+
+ chart_path = (
+ request.cls.result_dir
+ / f"{request.cls.csv_filename.replace('.csv', '')}_pipelineFPS.png"
+ )
+ fps_comparison_chart(chart_path, results, fps_key="Pipeline FPS (Video frames)")
+
+ chart_path = (
+ request.cls.result_dir
+ / f"{request.cls.csv_filename.replace('.csv', '')}_sfFPS_target.png"
+ )
+ fps_comparison_chart(
+ chart_path, results, fps_key="sf+roi+det FPS (Target frames)"
+ )
+
+ chart_path = (
+ request.cls.result_dir
+ / f"{request.cls.csv_filename.replace('.csv', '')}_pipelineFPS_target.png"
+ )
+ fps_comparison_chart(
+ chart_path, results, fps_key="Pipeline FPS (Target frames)"
+ )
+
+ chart_path = (
+ request.cls.result_dir
+ / f"{request.cls.csv_filename.replace('.csv', '')}_displayFPS_target.png"
+ )
+ fps_comparison_chart(chart_path, results, fps_key="Display FPS")
+
+
+@pytest.fixture(autouse=True)
+def each_test_setup(request):
+ test_class_self = request.instance
+ if torch.cuda.is_available():
+ torch.cuda.synchronize()
+
+ device = request.node.callspec.params.get("device")
+ detection_type = request.node.callspec.params.get("detection_type")
+ sf_enabled = request.node.callspec.params.get("sf_enabled")
+
+ test_class_self._testMethodName = (
+ f"sf_{detection_type}_{device}"
+ if sf_enabled
+ else f"yolo_{detection_type}_{device}"
+ )
+
+ # 1. Re-initialize a fresh configuration object
+ test_class_self.config = PipelineConfig(
+ CUSTOM_MODEL_FLAG=os.getenv("CUSTOM_MODEL_FLAG", CUSTOM_MODEL_FLAG_DEFAULT),
+ DEVICE=device.upper(),
+ OMIT_DETECTIONS_FLAG=True,
+ TEST_MODE=True,
+ DEBUG=os.getenv("DEBUG", DEBUG_DEFAULT),
+ DEBUG_FRAME_LIMIT=int(os.getenv("DEBUG_FRAME_LIMIT", 100)),
+ ENABLE_QUERYING=False,
+ MODEL_NAME=os.getenv("MODEL_NAME", MODEL_NAME_DEFAULT),
+ SMART_FILTERING_ENABLED=sf_enabled,
+ THRESHOLD_VALUE=int(os.getenv("THRESHOLD_VALUE", THRESHOLD_VALUE)),
+ DETECTION_TYPE=detection_type,
+ )
+
+ # video_output_name = f"{test_class_self._testMethodName}_detections_output.mp4"
+ vid_dir = test_class_self.result_dir / "results"
+ vid_dir.mkdir(parents=True, exist_ok=True)
+ # test_video_output_path = os.path.join(str(vid_dir), video_output_name)
+ os.environ["TEST_SUITE_RENDER_DIR"] = str(vid_dir)
+ test_class_self.config.SHARED_OUTPUT = str(test_class_self.result_dir)
+
+ # Reset core state machine properties before invoking any backend handlers
+ test_class_self.active = True
+ test_class_self._is_stopped = False
+
+ # FIXING ATTRIBUTEERRORS: Explicitly provision primitives before method re-binding
+ import threading
+ import time
+
+ test_class_self._stop_lock = threading.Lock()
+ # test_class_self.stat_start_time = time.perf_counter() # timing to display detection
+
+ test_class_self.device = test_class_self.config.DEVICE
+ test_class_self.device_input = test_class_self.config.device_input
+ test_class_self.resize_h, test_class_self.resize_w = [
+ test_class_self.config.MODEL_H,
+ test_class_self.config.MODEL_W,
+ ]
+
+ # Resolve concrete handler class type
+ HandlerClass = GPUStreamHandler if device == "gpu" else CPUStreamHandler
+ HandlerClass.pipeline_fn = test_class_self.__class__.pipeline_fn
+
+ # 2. Dynamically re-bind backend methods to this execution instance
+ import inspect
+ import types
+
+ handler_classes = [HandlerClass, DeviceBaseHandler]
+ all_method_names = set()
+
+ for cls in handler_classes:
+ for name, attr in inspect.getmembers(cls, predicate=inspect.isfunction):
+ all_method_names.add(name)
+
+ for method_name in all_method_names:
+ source_obj = (
+ HandlerClass if hasattr(HandlerClass, method_name) else DeviceBaseHandler
+ )
+ if hasattr(source_obj, method_name):
+ raw_func = getattr(source_obj, method_name)
+ if (
+ not hasattr(test_class_self.__class__, method_name)
+ or method_name == "pipeline_fn"
+ ):
+ setattr(
+ test_class_self,
+ method_name,
+ types.MethodType(raw_func, test_class_self),
+ )
+
+ # FIXING FILEEXISTSERRORS: Proactively clear lingering POSIX layout blocks in /dev/shm
+ from multiprocessing import shared_memory
+
+ for i in range(4): # Matches your self.ai_ring_depth footprint
+ stale_shm_name = f"shm_ai_640_{test_class_self.name}_{i}_{os.getpid()}"
+ try:
+ # Force link onto lingering handle and unlink it instantly from the OS map
+ lingering_shm = shared_memory.SharedMemory(name=stale_shm_name)
+ lingering_shm.close()
+ lingering_shm.unlink()
+ except FileNotFoundError:
+ pass
+
+ # 3. FORCE NATIVE PIPELINE PROVISIONING
+ test_class_self.setup_reader(
+ test_class_self.config.TARGET_FPS, test_class_self.config.CLIP_DURATION
+ )
+ test_class_self.initialize_variables()
+ test_class_self.setup_model(None)
+ test_class_self.prepare_pipeline()
+
+ # Reset runtime loop state properties
+ test_class_self.render_queue = None
+ test_class_self.frame_count_target = 0
+ test_class_self.next_process_idx = 0.0
+ test_class_self.frame_in_clip_count = 0
+ test_class_self.frame_count = 0
+ test_class_self.elapsed_display_time = 0.0
+
+ if hasattr(test_class_self, "setup_threads"):
+ test_class_self.setup_threads()
+
+ shared_event_buffer = {"sf": [], "roi": [], "det": []}
+ test_class_self.gpu_event_buffer = shared_event_buffer
+
+ if torch.cuda.is_available():
+ torch.cuda.empty_cache()
+ torch.cuda.ipc_collect()
+
+ yield
+
+ # Teardown logic
+ if (
+ hasattr(test_class_self, "render_queue")
+ and test_class_self.render_queue is not None
+ ):
+ try:
+ test_class_self.render_queue.put(None)
+ except Exception:
+ pass
+
+ render_proc_handle = getattr(test_class_self, "render_proc", None)
+ if render_proc_handle is not None and render_proc_handle._started.is_set():
+ test_class_self.render_proc.join(timeout=10.0)
+
+ # Signal the state machine to drop out of loops safely
+ test_class_self.active = False
+ if hasattr(test_class_self, "reader") and test_class_self.reader is not None:
+ try:
+ test_class_self.reader.stop()
+ except Exception:
+ pass
+
+ # Execute custom release blocks
+ test_class_self.stop()
+
+ # CRITICAL FIX: Block and wait for the background producer thread to die
+ # BEFORE we delete the reader attribute from the namespace map.
+ producer_thread_handle = getattr(test_class_self, "process_thread", None)
+ if producer_thread_handle is not None and producer_thread_handle.is_alive():
+ producer_thread_handle.join(timeout=5.0)
+
+ # Now it is structurally safe to purge attributes without causing cross-thread collisions
+ keys_to_purge = [
+ "reader",
+ "process_thread",
+ "inference_stream",
+ "bgs_stream",
+ "model",
+ "raw_input",
+ "ai_gpu_staging",
+ "ai_pinned_tensors",
+ "executor",
+ "io_executor",
+ ]
+ for key in keys_to_purge:
+ if hasattr(test_class_self, key):
+ delattr(test_class_self, key)
+
+ if (
+ hasattr(test_class_self, "gpu_event_buffer")
+ and test_class_self.gpu_event_buffer
+ ):
+ test_class_self.gpu_event_buffer.clear()
+ if hasattr(test_class_self, "component_stats") and test_class_self.component_stats:
+ test_class_self.component_stats.clear()
+
+ with torch.inference_mode():
+ if torch.cuda.is_available():
+ torch.cuda.synchronize()
+ torch.cuda.empty_cache()
+ torch.cuda.ipc_collect()
+
+ gc.collect()
+ time.sleep(0.2)
+
+
+@pytest.mark.usefixtures("setup_context")
+class TestSmartFilteringDetections:
+ # class TestSmartFilteringDetections(DeviceBaseHandler):
+ # """
+ # Unified testing harness.
+ # By inheriting from DeviceBaseHandler, 'self' functions as both
+ # the pytest telemetry harness and the live stream execution runner.
+ # """
+ # # Overriding __init__ to prevent standard initialization collisions with pytest
+ # def __init__(self, *args, **kwargs):
+ # pass
+ # SETUP --------------------------------------------
+
+ @pytest.mark.parametrize("device", ["cpu", "gpu"])
+ @pytest.mark.parametrize("sf_enabled", [True, False])
+ @pytest.mark.parametrize("detection_type", ["motion", "object"])
+ def test_detections(self, detection_type, device, sf_enabled):
+ """Unified test runner for all configurations."""
+ if detection_type == "motion" and not sf_enabled:
+ pytest.skip(
+ "Pure YOLO mode is structurally invalid for detection_type 'motion'."
+ )
+
+ # Run the actual model loader
+ # self.get_model_by_device(device, sf_enabled=self.config.sf_enabled)
+
+ # Execute
+ self.run_pipeline() # pipeline_fn)
+
+ def setup_threads(self):
+ # Shared 10MB memory for display
+ self.setup_shared_memory()
+
+ # Executor for Async YOLO tasks and FFmpeg re-encoding
+ # self.executor = ThreadPoolExecutor(max_workers=self.config.MAX_WORKERS)
+ # self.clip_executor = ThreadPoolExecutor(max_workers=self.config.MAX_WORKERS)
+
+ print(
+ f"sf_enabled: {self.config.sf_enabled}\tTEST_MODE: {self.config.TEST_MODE}",
+ flush=True,
+ )
+
+ # Producer: Handles acquisition and AI metadata logs
+ # self.process_thread = threading.Thread(
+ # target=self.run_pipeline,
+ # daemon=True,
+ # )
+
+ self.signal_queue = mp.Queue(maxsize=1)
+ self.render_queue = mp.Queue(maxsize=5)
+
+ # if self.config.TEST_MODE:
+ test_dir = os.getenv(
+ "TEST_SUITE_RENDER_DIR", str(Path(self.config.SHARED_OUTPUT))
+ )
+ os.makedirs(test_dir, exist_ok=True)
+
+ video_output_name = f"{self._testMethodName}_detections_output.mp4"
+ out_path = os.path.join(test_dir, video_output_name)
+
+ log_to_logger(
+ f"[TEST MODE] Detection results saved to: {out_path}", level="info"
+ )
+ self.render_proc = threading.Thread(
+ target=rendering_worker,
+ args=(
+ self.render_queue,
+ (self.disp_w, self.disp_h),
+ out_path,
+ self.target_fps,
+ ),
+ daemon=True,
+ )
+
+ # Dummy target alignment to prevent execution signature exceptions
+ self.display_proc = threading.Thread(target=lambda: None, daemon=True)
+ # else:
+ # self.render_proc = mp.Process(
+ # target=rendering_worker,
+ # args=(
+ # self.render_queue,
+ # self.shared_details,
+ # self.ready_buffer_idx,
+ # self.reader_active_idx,
+ # self.shm_frame_lengths,
+ # self.signal_queue,
+ # (self.disp_w, self.disp_h),
+ # self.config.DISPLAY_FRAME_QUALITY,
+ # ),
+ # )
+
+ # def display_signal_sync():
+ # while self.active:
+ # # Wait for signal
+ # # if self.mp_frame_ready_event.wait(timeout=1.0):
+ # # self.mp_frame_ready_event.clear()
+ # try:
+ # _ = self.signal_queue.get(timeout=1.0)
+ # # print(f"[DEBUG]: Signal received in FastAPI process for {self.name}", flush=True)
+ # # Wake FastAPI async loop in main thread
+ # self.loop.call_soon_threadsafe(self.frame_ready_event.set)
+ # except queue.Empty:
+ # continue
+
+ # self.display_proc = threading.Thread(
+ # target=display_signal_sync, daemon=True
+ # )
+
+ # if self.config.ENABLE_QUERYING:
+ # # NEW: Dedicated I/O pool for Disk/GPU transfers (Higher worker count for 8K)
+ # self.io_executor = ThreadPoolExecutor(max_workers=8)
+
+ # # Dedicated FFmpeg pool so re-encoding doesn't slow down live AI
+ # self.ffmpeg_executor = ThreadPoolExecutor(max_workers=2)
+
+ # if not self.config.TEST_MODE:
+ # # Sends metadata to VDMS
+ # self.metadata_thread = threading.Thread(
+ # target=send_metadata,
+ # args=(
+ # VDMSPool(self.config.DBHOST, self.config.DBPORT, size=10),
+ # self.config.DEBUG_FLAG,
+ # self.config.INGESTION,
+ # self.config.TEST_MODE,
+ # self.config.UDF_HOST,
+ # self.config.UDF_PORT,
+ # self.config.DBHOST,
+ # self.config.DBPORT,
+ # ),
+ # daemon=True,
+ # )
+
+ # # Consumer: Handles GPU-to-CPU download and Disk I/O (Writing resized frames to RAM disk)
+ # self.writer_thread = threading.Thread(
+ # target=self.video_writer_core_loop,
+ # args=(self.stop_writer,),
+ # daemon=True,
+ # )
+
+ # HELPERS --------------------------------------------
+ def _print_gpu_mem(self):
+ if torch.cuda.is_available():
+ # Memory currently used by tensors
+ allocated = torch.cuda.memory_allocated(0) / 1024**2
+ # Total memory reserved by PyTorch (the "Pool")
+ reserved = torch.cuda.memory_reserved(0) / 1024**2
+ print(f"\tAllocated: {allocated:0.2f} MB")
+ print(f"\tReserved: {reserved:0.2f} MB")
+ else:
+ print("\tCUDA not available.")
+
+ # def get_model_by_device(self, device, sf_enabled=False):
+ # """Singleton loader: only loads if device changes or model is missing."""
+ # if (
+ # sf_enabled
+ # and (self.frame_width * self.frame_height)
+ # <= self.config.SMART_FILTERING_PIXEL_CONSTRAINT
+ # ):
+ # sf_enabled = False
+
+ # if (
+ # TestSmartFilteringDetections._shared_model is not None
+ # and TestSmartFilteringDetections._shared_model_device == device
+ # and TestSmartFilteringDetections._shared_model_sf_enabled == sf_enabled
+ # ):
+ # self.model = TestSmartFilteringDetections._shared_model
+ # self.model_path = TestSmartFilteringDetections._shared_model_path
+ # return
+
+ # run_platform_name = "engine" if "cuda" in self.device_input else "openvino"
+
+ # if self.config.CUSTOM_MODEL_FLAG:
+ # dir_path = "/home/resources/models/ultralytics/custom_models"
+ # else:
+ # dir_path = f"/home/resources/models/ultralytics/{self.config.MODEL_NAME}/{self.config.MODEL_PRECISION}"
+
+ # (
+ # TestSmartFilteringDetections._shared_model,
+ # TestSmartFilteringDetections._shared_model_path,
+ # self.label_source,
+ # ) = get_model(
+ # # model_run_key, model_run_config, export=False
+ # Path(dir_path),
+ # self.config.MODEL_NAME,
+ # run_platform_name,
+ # self.device_input,
+ # batch=self.config.MODEL_MAX_BATCH_SIZE,
+ # force_export=force_export,
+ # sf_enabled=sf_enabled,
+ # model_h=self.resize_h,
+ # model_w=self.resize_w,
+ # )
+ # TestSmartFilteringDetections._shared_model_device = device
+ # TestSmartFilteringDetections._shared_model_sf_enabled = sf_enabled
+ # self.model = TestSmartFilteringDetections._shared_model
+ # # self.model.half()
+ # self.model_path = TestSmartFilteringDetections._shared_model_path
+
+ # W, H = self.resize_w, self.resize_h
+ # if not sf_enabled:
+ # W, H = self.frame_width, self.frame_height
+ # self.model_warmup(H, W)
+ # # self.pipeline_handler.model = self.model
+ # # self.pipeline_handler.model_warmup(H, W)
+
+ # def calculate_unique_coverage(self, merged_boxes, target_w=640, target_h=640):
+ # """
+ # FULLY VECTORIZED: Calculate pixel coverage without Python loops.
+ # Works for any number of boxes (1 to 1000+) with near-zero overhead.
+ # """
+ # if merged_boxes is None or merged_boxes.shape[0] == 0:
+ # return 0.0
+
+ # # 1. Scaling (Vectorized)
+ # # merged_boxes is [N, 4] -> [x1, y1, x2, y2] in 8K space
+ # scale = torch.tensor(
+ # [
+ # target_w / self.frame_width,
+ # target_h / self.frame_height,
+ # target_w / self.frame_width,
+ # target_h / self.frame_height,
+ # ],
+ # device=self.device_input,
+ # )
+
+ # # Scale and clamp all boxes at once on the GPU
+ # coords = (merged_boxes * scale).long()
+ # coords[:, [0, 2]] = coords[:, [0, 2]].clamp(0, target_w)
+ # coords[:, [1, 3]] = coords[:, [1, 3]].clamp(0, target_h)
+
+ # # 2. Vectorized Mask Filling
+ # # We create a 1D representation of the 640x640 mask for fast indexing
+ # mask = torch.zeros(
+ # target_h * target_w, device=self.device_input, dtype=torch.uint8
+ # )
+
+ # # For each box, we generate a range of indices and fill them.
+ # # Note: For small N (drones), a loop is okay, but for 'Swarm' noise,
+ # # we use this broadcasted approach:
+ # for i in range(coords.shape[0]):
+ # x1, y1, x2, y2 = coords[i]
+ # # Generate row indices for this box
+ # rows = torch.arange(y1, y2, device=self.device_input).view(-1, 1)
+ # # Calculate mask indices: (y * width) + x
+ # # This fills a horizontal slice of the mask in one GPU operation
+ # indices = (rows * target_w) + torch.arange(x1, x2, device=self.device_input)
+ # mask[indices] = 1
+
+ # # 3. Final Sum (GPU Reduction)
+ # return (torch.sum(mask).item() / (target_w * target_h)) * 100
+
+ def calculate_unique_coverage(self, merged_boxes, target_w=640, target_h=640):
+ """
+ TRUE LOOPLESS VECTORIZATION: Calculates combined bounding box pixel coverage
+ in a single GPU pass, completely avoiding CPU-GPU synchronization stalls.
+ """
+ if merged_boxes is None or merged_boxes.shape[0] == 0:
+ return 0.0
+
+ scale = torch.tensor(
+ [
+ target_w / self.frame_width,
+ target_h / self.frame_height,
+ target_w / self.frame_width,
+ target_h / self.frame_height,
+ ],
+ device=self.device_input,
+ )
+
+ coords = (merged_boxes * scale).long()
+ coords[:, [0, 2]] = coords[:, [0, 2]].clamp(0, target_w)
+ coords[:, [1, 3]] = coords[:, [1, 3]].clamp(0, target_h)
+
+ x1 = coords[:, 0].view(-1, 1, 1)
+ y1 = coords[:, 1].view(-1, 1, 1)
+ x2 = coords[:, 2].view(-1, 1, 1)
+ y2 = coords[:, 3].view(-1, 1, 1)
+
+ grid_y, grid_x = torch.meshgrid(
+ torch.arange(target_h, device=self.device_input),
+ torch.arange(target_w, device=self.device_input),
+ indexing="ij",
+ )
+
+ inside_boxes = (grid_x >= x1) & (grid_x < x2) & (grid_y >= y1) & (grid_y < y2)
+ unique_mask = torch.any(inside_boxes, dim=0)
+
+ return (torch.sum(unique_mask).item() / (target_w * target_h)) * 100
+
+ def _finalize_benchmarks(
+ self,
+ # n_frames,
+ num_objs,
+ total_pipeline_ms,
+ real_world_latency_ms,
+ coverage_percentages,
+ sf_enabled,
+ stat_frame_count,
+ stat_fps,
+ ):
+ """Aggregates metrics and adds them to the results list."""
+ latency_s = total_pipeline_ms / 1000.0 # Just pipeline (sf + roi_ det)
+ est_fps = self.frame_count / latency_s if latency_s > 0 else 0
+ est_fps_processed = self.frame_count_target / latency_s if latency_s > 0 else 0
+ duration_s = self.frame_count / self.input_fps if self.input_fps > 0 else 0
+
+ real_latency_s = real_world_latency_ms / 1000.0
+ real_est_fps = self.frame_count / real_latency_s if real_latency_s > 0 else 0
+ real_est_fps_processed = (
+ self.frame_count_target / real_latency_s if real_latency_s > 0 else 0
+ )
+
+ # Calculate averages for component breakdowns
+ avg_sf = (
+ sum(self.component_stats["sf"]) / len(self.component_stats["sf"])
+ if self.component_stats["sf"]
+ else 0
+ )
+ avg_roi = (
+ sum(self.component_stats["roi"]) / len(self.component_stats["roi"])
+ if self.component_stats["roi"]
+ else 0
+ )
+ avg_det = (
+ sum(self.component_stats["det"]) / len(self.component_stats["det"])
+ if self.component_stats["det"]
+ else 0
+ )
+ avg_cov = (
+ sum(coverage_percentages) / len(coverage_percentages)
+ if coverage_percentages
+ else (100.0 if not sf_enabled else 0)
+ )
+
+ avg_crops = (
+ sum(self.crops_per_frame_list) / len(self.crops_per_frame_list)
+ if self.crops_per_frame_list
+ else 0
+ )
+
+ # Total Latency Sum (SF + ROI + DET)
+ total_sum = avg_sf + avg_roi + avg_det
+
+ # Calculate how often we hit a high-motion cap (e.g., 20 crops)
+ capped_frames = sum(1 for c in self.crops_per_frame_list if c >= 20)
+ cap_rate = (
+ (capped_frames / len(self.crops_per_frame_list)) * 100
+ if self.crops_per_frame_list
+ else 0
+ )
+
+ self.__class__.benchmarks.append(
+ {
+ "Test Name": self._testMethodName,
+ "Detection Type": self.config.DETECTION_TYPE,
+ "Device": self.device,
+ "Smart Filtering": "Enabled" if sf_enabled else "Disabled",
+ "Video": self.video_path.name, # self.name?
+ "Video FPS": f"{self.input_fps:.2f}",
+ "Video Duration (s)": f"{duration_s:.4f}",
+ "Video Frames": self.frame_count,
+ "Target Frames": self.frame_count_target,
+ "Pipeline Latency (s)": f"{real_latency_s:.2f}",
+ "Display Latency (s)": f"{self.elapsed_display_time:.2f}",
+ "sf+roi+det Latency (s)": f"{latency_s:.2f}",
+ # Includes time by HW decoder (reader) and preparing output video
+ "Pipeline FPS (Video frames)": f"{real_est_fps:.2f}",
+ "Pipeline FPS (Target frames)": f"{real_est_fps_processed:.2f}",
+ # TIming to display frame (read to after send to render queue)
+ "Display Frames": stat_frame_count,
+ "Display FPS": f"{stat_fps:.2f}",
+ # Only sf + roi + det
+ "sf+roi+det FPS (Video frames)": f"{est_fps:.2f}",
+ "sf+roi+det FPS (Target frames)": f"{est_fps_processed:.2f}",
+ "Avg SF (ms)": f"{avg_sf:.2f}",
+ "Avg ROI (ms)": f"{avg_roi:.2f}",
+ "Avg Obj. Detection (ms)": f"{avg_det:.2f}",
+ "Total Breakdown Sum (ms)": f"{total_sum:.2f}",
+ "Avg Area Coverage %": f"{avg_cov:.2f}%",
+ "Avg Crops/Frame": f"{avg_crops:.1f}",
+ "Crop Cap Rate (>20)": f"{cap_rate:.1f}%",
+ "Objects Detected": num_objs,
+ }
+ )
+
+ print(f"\n[{self._testMethodName}] Latency: {latency_s:.2f} sec")
+ print(
+ f"\n[{self._testMethodName}] Pipeline FPS (Target frames): {real_est_fps_processed:.2f} ({self.frame_count_target} frames)"
+ )
+ print(
+ f"\n[{self._testMethodName}] sf+roi+det FPS (Target frames): {est_fps_processed:.2f} ({self.frame_count_target} frames)"
+ )
+ print(
+ f"\n[{self._testMethodName}] Display FPS (Target frames): {stat_fps:.2f} ({stat_frame_count} frames)"
+ )
+
+ # def tensor2opencv_gpu(self, frame_tensor):
+ # """
+ # GPU-native equivalent of tensor2opencv.
+ # Fixes the '3x3 ghosting' and 'ValueError' at 30+ FPS.
+ # """
+ # # 1. Handle Batch Dimension: (1, 3, 640, 640) -> (3, 640, 640)
+ # temp = frame_tensor.squeeze(0) if frame_tensor.ndim == 4 else frame_tensor
+
+ # # 2. Fix Layout & Shape: (3, 640, 640) -> (640, 640, 3)
+ # # contiguous() physically rearranges pixels in VRAM to interleave colors.
+ # # reshape() ensures we never see the (1, 409600, 3) shape again.
+ # gpu_hwc = temp.permute(1, 2, 0).reshape(640, 640, 3).contiguous()
+
+ # # 3. Visibility Fix: Scale floats (0.0-1.0) to bytes (0-255) on GPU
+ # if gpu_hwc.dtype != torch.uint8:
+ # gpu_hwc = (gpu_hwc * 255).clamp(0, 255).byte()
+
+ # # 4. Color Space Fix: RGB -> BGR (Matches OpenCV)
+ # gpu_bgr = gpu_hwc.flip(-1).contiguous()
+
+ # # 5. Bridge to CuPy (Zero-Copy)
+ # cp_frame = cp.from_dlpack(torch.utils.dlpack.to_dlpack(gpu_bgr))
+
+ # # 6. Select Double-Buffer from Pool
+ # target_buf_cpu = self.pinned_pool_np[self.pool_idx]
+ # target_buf_gpu = self.cp_pool[self.pool_idx]
+ # self.pool_idx = (self.pool_idx + 1) % 2
+
+ # if not hasattr(self, "transfer_stream"):
+ # self.transfer_stream = cp.cuda.Stream(non_blocking=True)
+
+ # with self.transfer_stream:
+ # # cp.copyto is more robust for pinned memory than target[:]
+ # cp.copyto(target_buf_gpu, cp_frame)
+ # # Download into the pinned CPU RAM
+ # target_buf_gpu.get(out=target_buf_cpu)
+
+ # # 7. Mandatory Sync: Wait for the DMA transfer to hit RAM
+ # self.transfer_stream.synchronize()
+
+ # # Return a snapshot copy so the worker has private memory for 30 FPS
+ # return target_buf_cpu.copy()
+
+ # DEBUG FUNCTIONS --------------------------------------------
+ def debug_save_mask(self, frame_source, frame_num, rois=None):
+ debug_dir = self.result_dir / "debug_mask" / self._testMethodName
+ debug_dir.mkdir(parents=True, exist_ok=True)
+
+ if frame_num > self.config.DEBUG_FRAME_LIMIT:
+ return
+
+ # Download or copy the data
+ if hasattr(frame_source, "download"):
+ img_cpu = frame_source.download()
+ elif torch.is_tensor(frame_source):
+ # .contiguous() fixes the horizontal "shredding"/static look
+ temp = frame_source.squeeze(0) if frame_source.ndim == 4 else frame_source
+ # img_cpu = temp.permute(1, 2, 0).contiguous().cpu().numpy()
+ if temp.ndim == 1:
+ temp = temp.view(self.resize_h, self.resize_w)
+
+ if temp.ndim == 3:
+ img_cpu = temp.permute(1, 2, 0).contiguous().cpu().numpy()
+ else:
+ img_cpu = temp.contiguous().cpu().numpy()
+ else:
+ # For numpy arrays (like your pinned memory), ensure memory is linear
+ img_cpu = np.ascontiguousarray(frame_source)
+
+ # Fix Visibility (Normalization)
+ # If float, scale to 0-255. If uint8, leave as is to avoid "neon" colors.
+ if img_cpu.dtype != np.uint8:
+ if img_cpu.max() <= 1.0:
+ img_cpu = (img_cpu * 255).clip(0, 255).astype(np.uint8)
+ else:
+ img_cpu = img_cpu.astype(np.uint8)
+
+ # Handle Color Space
+ # OpenCV imwrite expects BGR. If 3-channel (RGB), swap. If 1-channel, save as-is.
+ if len(img_cpu.shape) == 3:
+ # img_cpu = cv2.cvtColor(img_cpu, cv2.COLOR_RGB2BGR)
+ pass
+ else:
+ img_cpu = cv2.cvtColor(img_cpu, cv2.COLOR_GRAY2BGR)
+
+ # Draw 8K Boxes (Scaled down)
+ if rois is not None:
+ h_img, w_img = img_cpu.shape[:2]
+ scale_x = float(w_img) / self.frame_width
+ scale_y = float(h_img) / self.frame_height
+ boxes = rois.cpu().tolist() if torch.is_tensor(rois) else rois
+ for box in boxes:
+ x1, y1, x2, y2 = [
+ int(box[0] * scale_x),
+ int(box[1] * scale_y),
+ int(box[2] * scale_x),
+ int(box[3] * scale_y),
+ ]
+ cv2.rectangle(img_cpu, (x1, y1), (x2, y2), (0, 0, 255), 2)
+
+ # Save to disk
+ save_path = debug_dir / f"mask_{frame_num:04d}.jpg"
+ cv2.imwrite(str(save_path), img_cpu)
+
+ def debug_save_img_roi(self, frame_source, bbs_full_res, frame_num):
+ debug_dir = self.result_dir / "debug_analysis" / self._testMethodName
+ debug_dir.mkdir(parents=True, exist_ok=True)
+
+ if frame_num > self.config.DEBUG_FRAME_LIMIT:
+ return
+
+ # Download/Copy the frame
+ if torch.is_tensor(frame_source):
+ # .contiguous() is CRITICAL here to fix the "shredded" look
+ temp = frame_source.squeeze(0) if frame_source.ndim == 4 else frame_source
+ img_cpu = temp.permute(1, 2, 0).contiguous().cpu().numpy()
+ elif hasattr(frame_source, "download"):
+ img_cpu = frame_source.download()
+ else:
+ img_cpu = np.ascontiguousarray(frame_source)
+
+ # Fix Shape: restore spatial grid if flattened
+ if img_cpu.ndim == 3 and img_cpu.shape[0] == 1:
+ img_cpu = img_cpu.reshape((self.resize_h, self.resize_w, 3))
+
+ # Fix Visibility: ONLY multiply if it's actually floating point
+ # If uint8 is multiplied by 255, it wraps around and creates "neon" colors
+ if img_cpu.dtype != np.uint8:
+ if img_cpu.max() <= 1.0:
+ img_cpu = (img_cpu * 255).clip(0, 255).astype(np.uint8)
+ else:
+ img_cpu = img_cpu.astype(np.uint8)
+
+ # Color Space: Standardize to BGR for imwrite
+ if len(img_cpu.shape) != 3:
+ img_cpu = cv2.cvtColor(img_cpu, cv2.COLOR_GRAY2BGR)
+
+ # Draw 8K Boxes (Scaled down)
+ h_img, w_img = img_cpu.shape[:2]
+ scale_x = w_img / self.frame_width
+ scale_y = h_img / self.frame_height
+
+ if bbs_full_res is not None:
+ boxes = (
+ bbs_full_res.cpu().tolist()
+ if torch.is_tensor(bbs_full_res)
+ else bbs_full_res
+ )
+ for box in boxes:
+ x1, y1, x2, y2 = [
+ int(box[0] * scale_x),
+ int(box[1] * scale_y),
+ int(box[2] * scale_x),
+ int(box[3] * scale_y),
+ ]
+ cv2.rectangle(img_cpu, (x1, y1), (x2, y2), (0, 0, 255), 2)
+
+ cv2.imwrite(str(debug_dir / f"analysis_{frame_num:04d}.jpg"), img_cpu)
+
+ def debug_save_crops(self, cropped_batch, frame_num):
+ """Saves the first 5 crops of a batch to the results directory."""
+ debug_dir = self.result_dir / "debug_crops" / self._testMethodName
+ debug_dir.mkdir(parents=True, exist_ok=True)
+
+ # Only save for the first self.config.DEBUG_FRAME_LIMIT frames to avoid disk bloat
+ if frame_num > self.config.DEBUG_FRAME_LIMIT:
+ return
+
+ for i, crop in enumerate(cropped_batch[: self.config.DEBUG_FRAME_LIMIT]):
+ # Convert GPU Tensor [C, H, W] -> NumPy [H, W, C]
+ if torch.is_tensor(crop):
+ # Reverse normalization (* 255) and permute to BGR
+ img = (crop.squeeze(0).permute(1, 2, 0) * 255).byte().cpu().numpy()
+ else:
+ img = crop
+
+ cv2.imwrite(str(debug_dir / f"frame_{frame_num}_crop_{i}.jpg"), img)
+
+ def debug_save_img(self, frame_source, frame_num):
+ debug_dir = self.result_dir / "debug_test" / self._testMethodName
+ debug_dir.mkdir(parents=True, exist_ok=True)
+
+ if frame_num > self.config.DEBUG_FRAME_LIMIT:
+ return
+
+ # Download/Copy the frame
+ if torch.is_tensor(frame_source):
+ # .contiguous() is CRITICAL here to fix the "shredded" look
+ temp = frame_source.squeeze(0) if frame_source.ndim == 4 else frame_source
+ img_cpu = temp.permute(1, 2, 0).contiguous().cpu().numpy()
+ elif hasattr(frame_source, "download"):
+ img_cpu = frame_source.download()
+ else:
+ img_cpu = np.ascontiguousarray(frame_source)
+
+ # Fix Shape: restore spatial grid if flattened
+ if img_cpu.ndim == 3 and img_cpu.shape[0] == 1:
+ img_cpu = img_cpu.reshape((self.resize_h, self.resize_w, 3))
+
+ # Fix Visibility: ONLY multiply if it's actually floating point
+ # If uint8 is multiplied by 255, it wraps around and creates "neon" colors
+ if img_cpu.dtype != np.uint8:
+ if img_cpu.max() <= 1.0:
+ img_cpu = (img_cpu * 255).clip(0, 255).astype(np.uint8)
+ else:
+ img_cpu = img_cpu.astype(np.uint8)
+
+ # Color Space: Standardize to BGR for imwrite
+ if len(img_cpu.shape) == 3:
+ pass
+ else:
+ img_cpu = cv2.cvtColor(img_cpu, cv2.COLOR_GRAY2BGR)
+
+ cv2.imwrite(str(debug_dir / f"analysis_{frame_num:04d}.jpg"), img_cpu)
+
+ # # FRAME PROCESSORS --------------------------------------------
+ # def run_pipeline(self): #, pipeline_fn):
+ # # n_frames = 0
+ # num_objs = 0
+ # self.frame_count_target = 0
+ # self.next_process_idx = 0.0
+ # self.frame_in_clip_count = 0
+ # total_pipeline_time_ms = 0.0 # Track pure latency
+ # coverage_percentages = []
+ # self.component_stats = {"sf": [], "roi": [], "det": []}
+ # self.gpu_event_buffer = {"sf": [], "roi": [], "det": []}
+ # self.crops_per_frame_list = []
+
+ # # PRE-SYNC: Ensure GPU is idle before timing starts
+ # # if self.device_input == "cuda":
+ # # # Pre-allocate CUDA events for isolated GPU timing
+ # # # start_event = torch.cuda.Event(enable_timing=True)
+ # # # end_event = torch.cuda.Event(enable_timing=True)
+
+ # # if torch.cuda.is_available():
+ # # torch.cuda.synchronize()
+
+ # self.start()
+ # time.sleep(0.1)
+
+ # # start_time = time.perf_counter()
+ # total_session_start = time.perf_counter()
+
+ # while self.active: # and n_frames < 10*self.fps:
+ # # ret, frame = self.cap.read()
+ # device_frame, frame_num = self.reader.read()
+ # if device_frame is None:
+ # if self.reader.stopped:
+ # self.active = False
+ # break
+ # continue
+
+ # # n_frames += 1
+ # self.frame_count += 1
+ # is_target_frame = float(frame_num) >= self.next_process_idx
+ # nob = 0
+
+ # # if self.device_input == "cuda":
+ # # # start_eve nt.record()
+ # # nob, metrics = self.pipeline_fn(device_frame, frame_num, is_target_frame)
+ # # # end_event.record()
+ # # # torch.cuda.synchronize()
+ # # # total_pipeline_time_ms += start_event.elapsed_time(end_event)
+ # # else:
+ # # start_t = time.perf_counter()
+ # if self.device_input == "cuda":
+ # # Record the reader's current layout availability milestone
+ # curr_event = torch.cuda.Event()
+ # curr_event.record()
+
+ # # Instruct the isolated inference stream to wait for this specific frame context
+ # self.inference_stream.wait_event(curr_event)
+ # with torch.cuda.stream(self.inference_stream):
+ # nob, metrics = self.pipeline_fn(device_frame, frame_num, is_target_frame)
+ # # self.inference_stream.synchronize()
+
+ # else:
+ # nob, metrics = self.pipeline_fn(device_frame, frame_num, is_target_frame)
+ # # total_pipeline_time_ms += (time.perf_counter() - start_t) * 1000
+
+ # num_objs += nob
+
+ # if is_target_frame and metrics != {}:
+ # num_crops = len(metrics["bbs"]) if metrics["bbs"] is not None else 0
+ # self.crops_per_frame_list.append(num_crops)
+
+ # # self.component_stats["roi"].append(metrics["roi_time"])
+
+ # # if self.device_input != "cuda":
+ # if metrics.get("sf_time"):
+ # self.component_stats["sf"].append(metrics["sf_time"])
+ # if metrics.get("det_time"):
+ # self.component_stats["det"].append(metrics["det_time"])
+ # if metrics.get("roi_time"):
+ # self.component_stats["roi"].append(metrics["roi_time"])
+
+ # # Calculate coverage OUTSIDE the timed block to prevent interference
+ # if self.config.sf_enabled and metrics.get("bbs") is not None:
+ # cov = self.calculate_unique_coverage(metrics["bbs"])
+ # coverage_percentages.append(cov)
+
+ # # POST-SYNC: Ensure all GPU tasks finished before timing ends
+ # if self.device_input == "cuda":
+ # torch.cuda.synchronize()
+
+ # # ACTUAL speed including all overhead
+ # real_world_latency_ms = (time.perf_counter() - total_session_start) * 1000
+
+ # # if self.device_input == "cuda":
+ # # # Move GPU event timings into component_stats lists
+ # # for key in ["sf", "det", "roi"]:
+ # # for start, end in self.gpu_event_buffer[key]:
+ # # self.component_stats[key].append(start.elapsed_time(end))
+
+ # # Calculate Pure processing latency
+ # # This excludes: Reader, Tensor2OpenCV, Queueing, and Video Writing
+ # total_pipeline_time_ms = (
+ # sum(self.component_stats["sf"])
+ # + sum(self.component_stats["roi"])
+ # + sum(self.component_stats["det"])
+ # )
+
+ # # assert num_objs > 0
+ # if num_objs == 0:
+ # print(f" [WARNING] No objects detected for {self._testMethodName}")
+
+ # self._finalize_benchmarks(
+ # self.frame_count,
+ # num_objs,
+ # total_pipeline_time_ms,
+ # real_world_latency_ms,
+ # coverage_percentages,
+ # self.config.sf_enabled,
+ # self.stat_frame_count,
+ # self.stat_fps,
+ # )
+
+ def run_pipeline(self):
+ num_objs = 0
+ self.frame_count_target = 0
+ self.next_process_idx = 0.0
+ self.frame_in_clip_count = 0
+ total_pipeline_time_ms = 0.0
+ coverage_percentages = []
+ self.component_stats = {"sf": [], "roi": [], "det": []}
+ self.crops_per_frame_list = []
+
+ self.step_size = (
+ float(self.input_fps) / float(self.target_fps)
+ if hasattr(self, "target_fps")
+ else 1.0
+ )
+
+ self.start()
+ time.sleep(0.1)
+
+ total_session_start = time.perf_counter()
+
+ while self.active:
+ # 1. CAPTURE COMPLETE CYCLE OVERHEAD
+ # t_cycle_start = time.perf_counter()
+
+ device_frame, frame_num = self.reader.read()
+ if device_frame is None:
+ if self.reader is None or (
+ hasattr(self.reader, "stopped") and self.reader.stopped
+ ):
+ # --- CRITICAL PIPELINE DRAIN START ---
+ print(
+ "[INFO] Reader reached EOF. Draining asynchronous workers and VRAM queues..."
+ )
+
+ # 1. Thread Pool Flush: Force the CPU thread to block here until
+ # every single background pipeline task in the executor finishes.
+ if hasattr(self, "executor") and self.executor:
+ self.executor.shutdown(wait=True)
+
+ # 2. Render Queue Flush: If you utilize an asynchronous video frame
+ # saving worker thread, wait for its queue tasks to bottom out.
+ if hasattr(self, "render_queue") and self.render_queue:
+ while not self.render_queue.empty():
+ time.sleep(0.01)
+
+ # 3. GPU Hardware Flush: Force the GPU to completely finish
+ # all remaining background subtraction, crops, and YOLO operations.
+ if self.device_input == "cuda":
+ torch.cuda.synchronize()
+
+ self.active = False
+ break
+ continue
+
+ self.stat_start_time = time.perf_counter() # timing to display detection
+ self.frame_count += 1
+ # is_target_frame = float(frame_num) >= self.next_process_idx
+ # CRITICAL CADENCE DRIFT FIX: If input FPS equals target FPS, or if Smart Filtering
+ # is disabled (YOLO baseline), force is_target_frame to ALWAYS be True.
+ # This completely bypasses floating-point accumulation drift errors.
+ if (
+ not self.config.sf_enabled
+ or abs(float(self.input_fps) - float(self.target_fps)) < 0.01
+ ):
+ is_target_frame = True
+ else:
+ is_target_frame = float(frame_num) >= self.next_process_idx
+
+ # 2. DISPATCH WORKLOADS ASYNCHRONOUSLY
+ metrics = {}
+ if is_target_frame:
+ self.next_process_idx += self.step_size
+
+ if self.device_input == "cuda":
+ # Instantiate a lightweight hardware fence event object
+ curr_event = torch.cuda.Event(enable_timing=False)
+ # curr_event.record()
+ # Record the exact milestone on the default stream right after fetching the frame data
+ curr_event.record(torch.cuda.default_stream())
+ self.inference_stream.wait_event(curr_event)
+ with torch.cuda.stream(self.inference_stream):
+ # 1. Isolate the incoming image buffer canvas
+ isolated_device_frame = (
+ device_frame.clone()
+ if torch.is_tensor(device_frame)
+ else device_frame.copy()
+ )
+
+ # 2. RUN BACKGROUND SUBTRACTION IMMEDIATELY ON THE PRODUCER TIMELINE
+ # This guarantees that the mask matches this exact frame_num before threads overlap!
+ nob, metrics = self.pipeline_fn(
+ isolated_device_frame,
+ frame_num,
+ is_target_frame,
+ self.stat_start_time,
+ )
+ else:
+ # 1. Isolate the incoming image buffer canvas
+ isolated_device_frame = (
+ device_frame.clone()
+ if torch.is_tensor(device_frame)
+ else device_frame.copy()
+ )
+
+ # 2. RUN BACKGROUND SUBTRACTION IMMEDIATELY ON THE PRODUCER TIMELINE
+ # This guarantees that the mask matches this exact frame_num b
+ nob, metrics = self.pipeline_fn(
+ isolated_device_frame,
+ frame_num,
+ is_target_frame,
+ self.stat_start_time,
+ )
+
+ num_objs += nob
+ # else:
+ # # Process background execution context for skipped frames
+ # nob, metrics = self.pipeline_fn(
+ # device_frame, frame_num, is_target_frame, self.stat_start_time
+ # )
+
+ # 3. ENFORCE UNIFIED HARDWARE TIMING BARRIER
+ # if self.device_input == "cuda":
+ # torch.cuda.synchronize()
+
+ # t_cycle_end = time.perf_counter()
+ # cycle_total_ms = (t_cycle_end - t_cycle_start) * 1000.0
+
+ # 4. ALLOCATE ALL TRACKING TIMINGS ACCURATELY
+ # if is_target_frame:
+ if metrics != {}:
+ num_crops = len(metrics["bbs"]) if metrics["bbs"] is not None else 0
+ self.crops_per_frame_list.append(num_crops)
+
+ if self.config.sf_enabled:
+ self.component_stats["sf"].append(metrics["sf_time"])
+ self.component_stats["roi"].append(metrics["roi_time"])
+ else:
+ # Full-frame YOLO baseline accounts for total data movement cycle
+ # self.component_stats["det"].append(cycle_total_ms)
+ self.component_stats["sf"].append(0.0)
+ self.component_stats["roi"].append(0.0)
+
+ self.component_stats["det"].append(metrics["det_time"])
+
+ if self.config.sf_enabled and metrics.get("bbs") is not None:
+ cov = self.calculate_unique_coverage(metrics["bbs"])
+ coverage_percentages.append(cov)
+ # else:
+ # # CRITICAL METRICS FIX: If Smart Filtering runs on a skipped frame,
+ # # its mask generation overhead MUST be captured and tracked!
+ # if self.config.sf_enabled and metrics != {}:
+ # self.component_stats["sf"].append(metrics["sf_time"])
+
+ if self.device_input == "cuda":
+ torch.cuda.synchronize()
+
+ real_world_latency_ms = (time.perf_counter() - total_session_start) * 1000.0
+
+ total_pipeline_time_ms = (
+ sum(self.component_stats["sf"])
+ + sum(self.component_stats["roi"])
+ + sum(self.component_stats["det"])
+ )
+
+ if num_objs == 0:
+ print(f" [WARNING] No objects detected for {self._testMethodName}")
+
+ self._finalize_benchmarks(
+ # self.frame_count,
+ num_objs,
+ total_pipeline_time_ms,
+ real_world_latency_ms,
+ coverage_percentages,
+ self.config.sf_enabled,
+ self.stat_frame_count,
+ self.stat_fps,
+ )
+
+ # # TESTS --------------------------------------------
+ # def pipeline_fn(self, device_frame, overall_frame_num, is_target_frame):
+ # num_objs = 0
+ # metrics = {"sf_time": 0, "roi_time": 0, "det_time": 0, "bbs": None}
+
+ # # Initialize timing event handle pairs
+ # sf_start, sf_end = torch.cuda.Event(enable_timing=True), torch.cuda.Event(enable_timing=True)
+ # roi_start, roi_end = torch.cuda.Event(enable_timing=True), torch.cuda.Event(enable_timing=True)
+ # det_start, det_end = torch.cuda.Event(enable_timing=True), torch.cuda.Event(enable_timing=True)
+
+ # # --- 1. MOTION MASK GENERATION GATE ---
+ # if self.config.sf_enabled:
+ # if self.device_input == "cuda":
+ # sf_start.record(self.inference_stream)
+ # inf_data = self.rbtd_full_gpu(device_frame)
+ # sf_end.record(self.inference_stream)
+ # else:
+ # t_start = time.perf_counter()
+ # inf_data = self.rbtd_full_cpu(device_frame)
+ # metrics["sf_time"] = (time.perf_counter() - t_start) * 1000.0
+ # else:
+ # inf_data = {}
+
+ # # --- PIPELINE AT TARGET RATE ---
+ # if not is_target_frame:
+ # # If skipping, pull outstanding execution records immediately
+ # if self.device_input == "cuda" and self.config.sf_enabled:
+ # self.inference_stream.synchronize()
+ # metrics["sf_time"] = sf_start.elapsed_time(sf_end)
+ # return num_objs, metrics
+
+ # self.next_process_idx += self.step_size
+ # self.frame_count_target += 1
+ # self.frame_in_clip_count += 1
+ # inf_data["frameNum"] = self.frame_count_target
+
+ # # --- 2. FULL-RESOLUTION ROI EXTRACTION MAPS ---
+ # bbs_full_res = None
+ # if self.config.sf_enabled:
+ # if self.device_input == "cuda":
+ # roi_start.record(self.inference_stream)
+ # bbs_full_res = self.get_gpu_rois(
+ # inf_data["full_frame"],
+ # self.frame_count_target,
+ # inf_data["mask"],
+ # )
+ # roi_end.record(self.inference_stream)
+ # metrics["bbs"] = bbs_full_res
+ # else:
+ # t_start = time.perf_counter()
+ # bbs_full_res = self.get_cpu_rois(
+ # inf_data["full_frame"],
+ # self.frame_count_target,
+ # inf_data["mask"],
+ # )
+ # metrics["roi_time"] = (time.perf_counter() - t_start) * 1000.0
+ # metrics["bbs"] = bbs_full_res
+
+ # if self.config.DEBUG_FLAG:
+ # if self.device_input == "cuda":
+ # torch.cuda.stream(self.inference_stream)
+ # torch.cuda.synchronize()
+ # display_source = inf_data["mask"]
+ # self.debug_save_mask(
+ # display_source, self.frame_count_target, rois=bbs_full_res
+ # )
+
+ # clean_bbs = []
+ # if self.config.sf_enabled and bbs_full_res is not None:
+ # if torch.is_tensor(bbs_full_res):
+ # clean_bbs = bbs_full_res.detach().cpu().numpy()
+ # else:
+ # clean_bbs = np.array(bbs_full_res)
+
+ # # --- 3. MODEL INFERENCE TIMING BLOCK ---
+ # if self.device_input == "cuda":
+ # det_start.record(self.inference_stream)
+ # else:
+ # t_start = time.perf_counter()
+
+ # if self.config.DETECTION_TYPE != "motion":
+ # det_frame = inf_data["full_frame"] if "full_frame" in inf_data else device_frame
+ # merged = clean_bbs if self.config.sf_enabled else None
+ # metadata, _ = self.get_detections(
+ # det_frame,
+ # self.frame_in_clip_count,
+ # merged=merged,
+ # thickness=self.config.THICKNESS,
+ # device_input=self.config.device_input,
+ # )
+ # num_objs = len(metadata.keys())
+ # else:
+ # num_objs = len(clean_bbs)
+ # metadata = clean_bbs
+
+ # if self.device_input == "cuda":
+ # det_end.record(self.inference_stream)
+
+ # # CRITICAL PERFORMANCE SYNCHRONIZATION POINT:
+ # # We execute a single stream synchronization barrier here at the end of the entire loop.
+ # # This allows the GPU kernels to run overlapped and fully concurrently!
+ # self.inference_stream.synchronize()
+
+ # # Unpack hardware timings efficiently
+ # if self.config.sf_enabled:
+ # metrics["sf_time"] = sf_start.elapsed_time(sf_end)
+ # metrics["roi_time"] = roi_start.elapsed_time(roi_end)
+ # metrics["det_time"] = det_start.elapsed_time(det_end)
+ # metrics["bbs"] = bbs_full_res
+ # else:
+ # metrics["det_time"] = (time.perf_counter() - t_start) * 1000.0
+
+ # # --- 4. RENDER WORKER PREPARATION ---
+ # display_source = inf_data["full_frame"] if (inf_data and "full_frame" in inf_data) else device_frame
+
+ # if self.device_input == "cuda":
+ # gpu_resized = F.interpolate(
+ # display_source.unsqueeze(0).float(),
+ # size=(self.disp_h, self.disp_w),
+ # mode="bilinear",
+ # align_corners=False,
+ # ).squeeze(0).contiguous()
+ # disp_frame = np.copy(tensor2opencv(gpu_resized, self.config.device_input, is_bgr=True))
+ # else:
+ # cpu_resized = cv2.resize(device_frame, (self.disp_w, self.disp_h))
+ # disp_frame = np.copy(tensor2opencv(cpu_resized, self.config.device_input, is_bgr=True))
+
+ # data_to_draw = clean_bbs if self.config.DETECTION_TYPE == "motion" else metadata
+
+ # try:
+ # if hasattr(self, "render_queue") and getattr(self, "render_queue", None) is not None and not self.render_queue.full():
+ # self.render_queue.put(
+ # (
+ # disp_frame,
+ # inf_data["frameNum"] if "frameNum" in inf_data else self.frame_count_target,
+ # data_to_draw,
+ # self.label_source,
+ # )
+ # )
+ # if self.config.DEBUG_FLAG:
+ # self.debug_save_img(disp_frame, self.frame_count_target)
+ # self.debug_save_img_roi(disp_frame, bbs_full_res, self.frame_count_target)
+ # except queue.Full:
+ # pass
+
+ # self.update_frame()
+ # return num_objs, metrics
+
+ def pipeline_fn(
+ self, device_frame, overall_frame_num, is_target_frame, stat_start_time
+ ):
+ num_objs = 0
+ metrics = {"sf_time": 0, "roi_time": 0, "det_time": 0, "bbs": None}
+
+ # Pre-allocate non-blocking event records to extract clean GPU timings
+ sf_start, sf_end = (
+ torch.cuda.Event(enable_timing=True),
+ torch.cuda.Event(enable_timing=True),
+ )
+ roi_start, roi_end = (
+ torch.cuda.Event(enable_timing=True),
+ torch.cuda.Event(enable_timing=True),
+ )
+ det_start, det_end = (
+ torch.cuda.Event(enable_timing=True),
+ torch.cuda.Event(enable_timing=True),
+ )
+
+ # --- 1. MOTION MASK GENERATION GATE ---
+ if self.config.sf_enabled:
+ if self.device_input == "cuda":
+ sf_start.record(self.inference_stream)
+ inf_data = self.rbtd_full_gpu(device_frame)
+ sf_end.record(self.inference_stream)
+ else:
+ t_start = time.perf_counter()
+ inf_data = self.rbtd_full_cpu(device_frame)
+ metrics["sf_time"] = (time.perf_counter() - t_start) * 1000.0
+ else:
+ inf_data = {}
+
+ # --- PIPELINE AT TARGET RATE ---
+ if not is_target_frame:
+ # Safe, non-blocking timing extraction for skipped frames
+ if self.device_input == "cuda" and self.config.sf_enabled:
+ torch.cuda.synchronize()
+ metrics["sf_time"] = sf_start.elapsed_time(sf_end)
+ return num_objs, metrics
+
+ self.frame_count_target += 1
+ self.frame_in_clip_count += 1
+ inf_data["frameNum"] = self.frame_count_target
+
+ # --- 2. FULL-RESOLUTION ROI EXTRACTION MAPS ---
+ bbs_full_res = None
+ if self.config.sf_enabled:
+ if self.device_input == "cuda":
+ roi_start.record(self.inference_stream)
+ bbs_full_res = self.get_gpu_rois(
+ inf_data["full_frame"],
+ self.frame_count_target,
+ inf_data["mask"],
+ )
+ roi_end.record(self.inference_stream)
+ metrics["bbs"] = bbs_full_res
+ else:
+ t_start = time.perf_counter()
+ bbs_full_res = self.get_cpu_rois(
+ inf_data["full_frame"],
+ self.frame_count_target,
+ inf_data["mask"],
+ )
+ metrics["roi_time"] = (time.perf_counter() - t_start) * 1000.0
+ metrics["bbs"] = bbs_full_res
+
+ if self.config.DEBUG_FLAG:
+ # if self.device_input == "cuda":
+ # # Isolate data capturing using an asynchronous memory clone operation
+ # # This safely copies data without dropping your multi-stream execution pipeline concurrency
+ # display_source = inf_data["mask"].clone().to("cpu", non_blocking=True)
+ # else:
+ display_source = inf_data["mask"]
+ # self.inference_stream.synchronize()
+ # display_source = inf_data["mask"]
+ self.debug_save_mask(
+ display_source, self.frame_count_target, rois=bbs_full_res
+ )
+
+ clean_bbs = []
+ if self.config.sf_enabled and bbs_full_res is not None:
+ if torch.is_tensor(bbs_full_res):
+ clean_bbs = bbs_full_res.detach().cpu().numpy()
+ else:
+ clean_bbs = np.array(bbs_full_res)
+
+ # --- 3. MODEL INFERENCE TIMING BLOCK ---
+ if self.device_input == "cuda":
+ det_start.record(self.inference_stream)
+ else:
+ t_start = time.perf_counter()
+
+ if self.config.DETECTION_TYPE != "motion":
+ # det_frame = (
+ # inf_data["full_frame"] if "full_frame" in inf_data else device_frame
+ # )
+ # Isolate your image buffer array view to prevent upstream reader pointer races
+ if "full_frame" in inf_data:
+ det_frame = inf_data["full_frame"]
+ else:
+ det_frame = (
+ device_frame.clone()
+ if torch.is_tensor(device_frame)
+ else device_frame.copy()
+ )
+
+ merged = clean_bbs if self.config.sf_enabled else None
+ metadata, _ = self.get_detections(
+ det_frame,
+ self.frame_in_clip_count,
+ merged=merged,
+ thickness=self.config.THICKNESS,
+ device_input=self.config.device_input,
+ )
+ num_objs = len(metadata.keys())
+ else:
+ num_objs = len(clean_bbs)
+ metadata = clean_bbs
+
+ if self.device_input == "cuda":
+ det_end.record(self.inference_stream)
+
+ torch.cuda.synchronize()
+
+ # Extract hardware timings smoothly without stalling mid-run
+ if self.config.sf_enabled:
+ metrics["sf_time"] = sf_start.elapsed_time(sf_end)
+ metrics["roi_time"] = roi_start.elapsed_time(roi_end)
+ metrics["det_time"] = det_start.elapsed_time(det_end)
+ metrics["bbs"] = bbs_full_res
+ else:
+ metrics["det_time"] = (time.perf_counter() - t_start) * 1000.0
+
+ # --- 4. RENDER WORKER PREPARATION ---
+ display_source = (
+ inf_data["full_frame"]
+ if (inf_data and "full_frame" in inf_data)
+ else device_frame
+ )
+
+ if self.device_input == "cuda":
+ gpu_resized = (
+ F.interpolate(
+ display_source.unsqueeze(0).float(),
+ size=(self.disp_h, self.disp_w),
+ mode="bilinear",
+ align_corners=False,
+ )
+ .squeeze(0)
+ .contiguous()
+ )
+ disp_frame = np.copy(
+ tensor2opencv(gpu_resized, self.config.device_input, is_bgr=True)
+ )
+ else:
+ cpu_resized = cv2.resize(device_frame, (self.disp_w, self.disp_h))
+ disp_frame = np.copy(
+ tensor2opencv(cpu_resized, self.config.device_input, is_bgr=True)
+ )
+
+ data_to_draw = clean_bbs if self.config.DETECTION_TYPE == "motion" else metadata
+
+ # try:
+ # if (
+ # hasattr(self, "render_queue")
+ # and getattr(self, "render_queue", None) is not None
+ # and not self.render_queue.full()
+ # ):
+ # self.render_queue.put(
+ # (
+ # disp_frame,
+ # inf_data["frameNum"]
+ # if "frameNum" in inf_data
+ # else self.frame_count_target,
+ # data_to_draw,
+ # self.label_source,
+ # )
+ # )
+ # if self.config.DEBUG_FLAG:
+ # self.debug_save_img(disp_frame, self.frame_count_target)
+ # self.debug_save_img_roi(
+ # disp_frame, bbs_full_res, self.frame_count_target
+ # )
+ # except queue.Full:
+ # pass
+
+ if (
+ hasattr(self, "render_queue")
+ and getattr(self, "render_queue", None) is not None
+ ):
+ self.render_queue.put(
+ (
+ disp_frame,
+ inf_data["frameNum"]
+ if "frameNum" in inf_data
+ else self.frame_count_target,
+ data_to_draw,
+ self.label_source,
+ )
+ )
+ if self.config.DEBUG_FLAG:
+ self.debug_save_img(disp_frame, self.frame_count_target)
+ self.debug_save_img_roi(disp_frame, bbs_full_res, self.frame_count_target)
+
+ self.update_frame(stat_start_time)
+ return num_objs, metrics
+
+ # def sf_cpu_pipeline_fn(self, device_frame, overall_frame_num, is_target_frame):
+ # num_objs = 0
+ # metrics = {"sf_time": 0, "roi_time": 0, "det_time": 0, "bbs": None}
+
+ # # Smart Filtering (Resize / Background Subtraction / Threshold / Dilate)
+ # t_start = time.perf_counter()
+ # inf_data = self.rbtd_full_cpu(device_frame)
+ # metrics["sf_time"] = (time.perf_counter() - t_start) * 1000
+
+ # # Only keep frames for TARGET_FPS
+ # # Videos are written for target frames only
+ # if is_target_frame:
+ # self.next_process_idx += self.step_size
+ # self.frame_count_target += 1 # 1-indexed
+
+ # if not inf_data or inf_data["mask"] is None:
+ # return num_objs, metrics
+
+ # metadata = {}
+ # bbs_to_send = []
+ # data_to_draw = []
+ # if inf_data:
+ # inf_data["frameNum"] = self.frame_count_target
+
+ # # Get ROIs
+ # t_start = time.perf_counter()
+ # bbs_full_res = self.get_cpu_rois(
+ # inf_data["full_frame"],
+ # self.frame_count_target,
+ # inf_data["mask"],
+ # )
+ # metrics["roi_time"] = (time.perf_counter() - t_start) * 1000
+ # metrics["bbs"] = bbs_full_res
+
+ # if self.config.DEBUG_FLAG:
+ # display_source = inf_data["mask"]
+ # self.debug_save_mask(
+ # display_source, self.frame_count_target, rois=bbs_full_res
+ # )
+
+ # if bbs_full_res is not None and len(bbs_full_res) == 0:
+ # return num_objs, metrics
+
+ # # if self.config.DEBUG_FLAG or self.config.DETECTION_TYPE == "motion":
+ # # display_source = (
+ # # inf_data["full_frame"]
+ # # if (inf_data and "full_frame" in inf_data)
+ # # else device_frame
+ # # )
+ # # cpu_resized = cv2.resize(
+ # # display_source,
+ # # (self.resize_w, self.resize_h),
+ # # interpolation=cv2.INTER_NEAREST,
+ # # )
+
+ # t_start = time.perf_counter()
+ # if self.config.DETECTION_TYPE == "motion":
+ # # Motion Mode: Prepare boxes for drawing
+ # if (
+ # bbs_full_res is not None
+ # and bbs_full_res.ndim == 2
+ # and bbs_full_res.size(0) > 0
+ # ):
+ # scaled_resized_bbs = bbs_full_res / self.scales_tensor
+ # bbs_to_send = scaled_resized_bbs.cpu().tolist()
+ # else:
+ # bbs_to_send = []
+
+ # data_to_draw = bbs_to_send
+ # num_objs = len(bbs_full_res)
+ # else:
+ # # Object Mode: Run YOLO and prepare metadata
+ # det_frame = inf_data["full_frame"] if inf_data else device_frame
+ # metadata, _ = self.get_detections(
+ # det_frame,
+ # self.frame_count_target, # Frame used in metadata
+ # merged=bbs_full_res,
+ # thickness=self.config.THICKNESS,
+ # device_input=self.config.device_input,
+ # )
+ # data_to_draw = metadata
+ # num_objs = len(metadata.keys())
+
+ # metrics["det_time"] = (time.perf_counter() - t_start) * 1000
+
+ # # --- OFFLOAD TO QUEUE (Run for EVERY target frame) ---
+ # if not self.config.TEST_MODE:
+ # display_source = (
+ # inf_data["full_frame"]
+ # if (inf_data and "full_frame" in inf_data)
+ # else device_frame
+ # )
+ # cpu_resized = cv2.resize(display_source, (self.resize_w, self.resize_h))
+ # display_frame = tensor2opencv(
+ # cpu_resized, self.config.device_input, is_bgr=True
+ # )
+ # self.render_queue.put((display_frame, data_to_draw, self.label_source))
+
+ # if self.config.DEBUG_FLAG:
+ # cpu_resized = cv2.resize(
+ # display_source, (self.resize_w, self.resize_h)
+ # )
+ # self.debug_save_img(cpu_resized, self.frame_count_target)
+ # self.debug_save_img_roi(
+ # cpu_resized, bbs_full_res, self.frame_count_target
+ # )
+
+ # return num_objs, metrics
+
+ # def sf_gpu_pipeline_fn(self, device_frame, overall_frame_num, is_target_frame):
+ # num_objs = 0
+ # metrics = {"sf_time": 0, "roi_time": 0, "det_time": 0, "bbs": None}
+
+ # # Smart Filtering (Resize / Background Subtraction / Threshold / Dilate)
+ # sf_start, sf_end = (
+ # torch.cuda.Event(enable_timing=True),
+ # torch.cuda.Event(enable_timing=True),
+ # )
+ # sf_start.record()
+ # inf_data = self.rbtd_full_gpu(device_frame)
+ # sf_end.record()
+ # self.gpu_event_buffer["sf"].append((sf_start, sf_end))
+
+ # # Only keep frames for TARGET_FPS
+ # # Videos are written for target frames only
+ # if is_target_frame:
+ # self.next_process_idx += self.step_size
+ # self.frame_count_target += 1 # 1-indexed
+
+ # if not inf_data or inf_data["mask"] is None:
+ # return num_objs, metrics
+
+ # metadata = {}
+ # bbs_to_send = []
+ # data_to_draw = []
+ # if inf_data:
+ # inf_data["frameNum"] = self.frame_count_target
+
+ # # Get ROIs
+ # # time.perf_counter accurate since called self.bgs_stream.waitForCompletion()
+ # # t_start = time.perf_counter()
+ # roi_start = torch.cuda.Event(enable_timing=True)
+ # roi_end = torch.cuda.Event(enable_timing=True)
+ # roi_start.record()
+ # bbs_full_res = self.get_gpu_rois(
+ # inf_data["full_frame"],
+ # self.frame_count_target,
+ # inf_data["mask"],
+ # )
+ # # metrics["roi_time"] = (time.perf_counter() - t_start) * 1000
+ # # self.gpu_event_buffer["roi"].append((roi_start, roi_end))
+ # roi_end.record()
+ # self.gpu_event_buffer["roi"].append((roi_start, roi_end))
+ # metrics["bbs"] = bbs_full_res
+
+ # if self.config.DEBUG_FLAG:
+ # self.bgs_stream.waitForCompletion()
+ # display_source = inf_data["mask"]
+ # self.debug_save_mask(
+ # display_source, self.frame_count_target, rois=bbs_full_res
+ # )
+
+ # if bbs_full_res is not None and len(bbs_full_res) == 0:
+ # return num_objs, metrics
+
+ # det_start, det_end = (
+ # torch.cuda.Event(enable_timing=True),
+ # torch.cuda.Event(enable_timing=True),
+ # )
+ # det_start.record()
+ # if self.config.DETECTION_TYPE == "motion":
+ # # Motion Mode: Prepare boxes for drawing
+ # if (
+ # bbs_full_res is not None
+ # and bbs_full_res.ndim == 2
+ # and bbs_full_res.size(0) > 0
+ # ):
+ # scaled_resized_bbs = bbs_full_res / self.scales_tensor
+ # bbs_to_send = scaled_resized_bbs.detach().cpu().tolist()
+ # else:
+ # bbs_to_send = []
+ # data_to_draw = bbs_to_send
+ # num_objs = len(bbs_to_send)
+ # else:
+ # # Object Mode: Run YOLO and prepare metadata
+ # det_frame = (
+ # inf_data["full_frame"] if inf_data else device_frame
+ # ) # RGB
+ # metadata, _ = self.get_detections(
+ # det_frame,
+ # self.frame_count_target,
+ # merged=bbs_full_res,
+ # thickness=self.config.THICKNESS,
+ # device_input=self.config.device_input,
+ # )
+ # data_to_draw = metadata
+ # num_objs = len(metadata.keys())
+
+ # det_end.record()
+ # self.gpu_event_buffer["det"].append((det_start, det_end))
+
+ # # --- OFFLOAD TO QUEUE (Run for EVERY target frame) ---
+ # if not self.config.TEST_MODE:
+ # display_source = (
+ # inf_data["full_frame"]
+ # if (inf_data and "full_frame" in inf_data)
+ # else device_frame
+ # ) # RGB
+ # gpu_resized = F.interpolate(
+ # display_source.unsqueeze(0).half(),
+ # size=(self.resize_h, self.resize_w),
+ # mode="bilinear",
+ # align_corners=False,
+ # ).squeeze(0)
+ # display_frame = tensor2opencv(
+ # gpu_resized, self.config.device_input, is_bgr=True
+ # )
+ # # display_frame = self.tensor2opencv_gpu(gpu_resized)
+ # self.render_queue.put((display_frame, data_to_draw, self.label_source))
+
+ # if self.config.DEBUG_FLAG:
+ # gpu_resized = F.interpolate(
+ # display_source.unsqueeze(0).float(),
+ # size=(self.resize_h, self.resize_w),
+ # mode="bilinear",
+ # align_corners=False,
+ # ) # RGB
+ # self.debug_save_img(gpu_resized, self.frame_count_target)
+ # self.debug_save_img_roi(
+ # gpu_resized, bbs_full_res, self.frame_count_target
+ # )
+
+ # # Async metrics stay at 0 for fairness; collected at the end of the video
+ # metrics["sf_time"] = 0
+ # metrics["roi_time"] = 0
+ # metrics["det_time"] = 0
+ # return num_objs, metrics
+
+ # def yolo_cpu_pipeline_fn(self, device_frame, overall_frame_num, is_target_frame):
+ # num_objs = 0
+ # metadata = {}
+ # metrics = {"sf_time": 0, "roi_time": 0, "det_time": 0, "bbs": None}
+
+ # # Only keep frames for TARGET_FPS
+ # if is_target_frame:
+ # self.next_process_idx += self.step_size
+ # self.frame_count_target += 1 # 1-indexed
+
+ # # Get detection at original resolution (No SF)
+ # t_start = time.perf_counter()
+ # metadata, _ = self.get_detections(
+ # device_frame,
+ # self.frame_count_target,
+ # thickness=self.config.THICKNESS,
+ # device_input=self.config.device_input,
+ # )
+ # num_objs = len(metadata.keys())
+ # metrics["det_time"] = (time.perf_counter() - t_start) * 1000
+
+ # if not self.config.TEST_MODE:
+ # cpu_resized = cv2.resize(device_frame, (self.resize_w, self.resize_h))
+ # display_frame = tensor2opencv(
+ # cpu_resized, self.config.device_input, is_bgr=True
+ # )
+ # self.render_queue.put((display_frame, metadata, self.label_source))
+
+ # return num_objs, metrics
+
+ # def yolo_gpu_pipeline_fn(self, frame_raw, overall_frame_num, is_target_frame):
+ # num_objs = 0
+ # metadata = {}
+ # metrics = {"sf_time": 0, "roi_time": 0, "det_time": 0, "bbs": None}
+
+ # # Only keep frames for TARGET_FPS
+ # if is_target_frame:
+ # self.next_process_idx += self.step_size
+ # self.frame_count_target += 1 # 1-indexed
+
+ # # Get detection at original resolution (No SF)
+ # det_start, det_end = (
+ # torch.cuda.Event(enable_timing=True),
+ # torch.cuda.Event(enable_timing=True),
+ # )
+ # det_start.record()
+ # metadata, _ = self.get_detections(
+ # frame_raw,
+ # self.frame_count_target,
+ # thickness=self.config.THICKNESS,
+ # device_input=self.config.device_input,
+ # )
+ # num_objs = len(metadata.keys())
+ # det_end.record()
+ # self.gpu_event_buffer["det"].append((det_start, det_end))
+
+ # # --- OFFLOAD TO QUEUE (Run for EVERY target frame) ---
+ # if not self.config.TEST_MODE:
+ # gpu_resized = F.interpolate(
+ # frame_raw.unsqueeze(0).half(),
+ # size=(self.resize_h, self.resize_w),
+ # mode="bilinear",
+ # align_corners=False,
+ # ).squeeze(0)
+ # display_frame = tensor2opencv(
+ # gpu_resized, self.config.device_input, is_bgr=True
+ # )
+ # # display_frame = self.tensor2opencv_gpu(gpu_resized)
+ # self.render_queue.put((display_frame, metadata, self.label_source))
+
+ # # Async metrics stay at 0 for fairness; collected at the end of the video
+ # metrics["sf_time"] = 0
+ # metrics["det_time"] = 0
+ # return num_objs, metrics
+
+
+# INHERIT METHODS FROM HANDLERS -----------------------------------------------------------------
+# TestSmartFilteringDetections.setup_reader = DeviceBaseHandler.setup_reader
+# TestSmartFilteringDetections.get_frameWH = DeviceBaseHandler.get_frameWH
+# TestSmartFilteringDetections.initialize_variables = (
+# DeviceBaseHandler.initialize_variables
+# )
+# TestSmartFilteringDetections.filter_contained_boxes = (
+# DeviceBaseHandler.filter_contained_boxes
+# )
+# TestSmartFilteringDetections.get_detections = DeviceBaseHandler.get_detections
+# TestSmartFilteringDetections.prepare_pipeline = DeviceBaseHandler.prepare_pipeline
+# TestSmartFilteringDetections.get_gpu_rois_by_area = (
+# DeviceBaseHandler.get_gpu_rois_by_area
+# )
+# TestSmartFilteringDetections.get_gpu_rois = DeviceBaseHandler.get_gpu_rois
+# TestSmartFilteringDetections.get_cpu_rois = DeviceBaseHandler.get_cpu_rois
+# TestSmartFilteringDetections.run_model = DeviceBaseHandler.run_model
+# TestSmartFilteringDetections.model_warmup = DeviceBaseHandler.model_warmup
+
+# # TestSmartFilteringDetections.cleanup_gpu = GPUStreamHandler.cleanup_gpu
+# # TestSmartFilteringDetections.rbtd_full_gpu = GPUStreamHandler.rbtd_full_gpu
+# # TestSmartFilteringDetections.prepare_gpu_pipeline = (
+# # GPUStreamHandler.prepare_gpu_pipeline
+# # )
+# # TestSmartFilteringDetections.allocate_gpu = GPUStreamHandler.allocate_gpu
+# TestSmartFilteringDetections.gpu_warmup = GPUStreamHandler.gpu_warmup
+# TestSmartFilteringDetections.apply_background_subtraction_gpu = (
+# GPUStreamHandler.apply_background_subtraction_gpu
+# )
+
+# TestSmartFilteringDetections.cleanup_cpu = CPUStreamHandler.cleanup_cpu
+# # TestSmartFilteringDetections.rbtd_full_cpu = CPUStreamHandler.rbtd_full_cpu
+# # TestSmartFilteringDetections.prepare_cpu_pipeline = (
+# # CPUStreamHandler.prepare_cpu_pipeline
+# # )
+# # TestSmartFilteringDetections.allocate_cpu = CPUStreamHandler.allocate_cpu
+# TestSmartFilteringDetections.apply_background_subtraction_cpu = (
+# CPUStreamHandler.apply_background_subtraction_cpu
+# )
+
+
+def get_pytest_filter_expression(args):
+ print("\n" + "=" * 50)
+ print("TARGET SELECTION PREVIEW")
+ print("=" * 50)
+
+ filter_expression = "test_detections"
+ # applied_subs = []
+
+ if args.sf_enabled is not None:
+ # Target exact parameter tokens generated by pytest parametrization
+ sf_str = "-True-" if args.sf_enabled else "-False-"
+ filter_expression += f" and {sf_str}"
+ # applied_subs.append(f"sf_enabled={args.sf_enabled}")
+
+ if args.detection_type:
+ filter_expression += f" and {args.detection_type}"
+
+ # Target hardware context selection filter applies globally across all test cases
+ if args.device.lower() != "all":
+ print(f" 💻 Hardware Context Constraint: {args.device.upper()}")
+ filter_expression = f"({filter_expression}) and {args.device.lower()}"
+ else:
+ print(" 💻 Hardware Context Constraint: ALL AVAILABLE")
+
+ print("=" * 50)
+ print(f"COMPILED PYTEST KEYWORD EXPRESSION:\n 👉 {filter_expression}")
+ print("=" * 50 + "\n")
+
+ return filter_expression
+
+
+if __name__ == "__main__":
+ # TEST ARGUMENTS
+ parser = argparse.ArgumentParser(description="Run Video Detection Pipeline Tests")
+ parser.add_argument(
+ "-s",
+ "--source",
+ type=str,
+ default="anduril_swarm_8K.mp4",
+ help="Video filename (located in /inputs)",
+ )
+
+ # MODEL TO USE
+ parser.add_argument(
+ "--no-custom",
+ action="store_false",
+ dest="custom_model_flag",
+ help="Enable if using Ultralytics YOLO model",
+ )
+ parser.add_argument(
+ "-m",
+ "--model",
+ type=str,
+ default="drone_detection",
+ dest="model_name",
+ help="Name of model. Required if `--no-custom` is enabled. [Default: drone_detection]",
+ )
+
+ # Filter tests
+ parser.add_argument(
+ "--type",
+ type=str,
+ choices=["object", "motion"],
+ dest="detection_type",
+ help="Filter by detection type (object or motion)",
+ )
+ parser.add_argument(
+ "--device",
+ type=str,
+ default="all",
+ choices=["cpu", "gpu", "all"],
+ help="Filter by device (cpu or gpu)",
+ )
+ parser.add_argument(
+ "--sf",
+ action="store_true",
+ default=None,
+ dest="sf_enabled",
+ help="Filter by Smart Filtering",
+ )
+
+ # DEBUGGING
+ parser.add_argument(
+ "--debug",
+ action="store_true",
+ help="Enable debug message and save intermediate images for Smart Filtering tests",
+ )
+ parser.add_argument(
+ "-n",
+ type=int,
+ default=100,
+ dest="debug_frame_limit",
+ help="Number of frames used for debugging [Default: 100]",
+ )
+
+ args = parser.parse_args()
+
+ # UPDATE ENVIRONMENTAL VARIABLES
+ os.environ["VIDEO_FILENAME"] = args.source
+ os.environ["CUSTOM_MODEL_FLAG"] = "True" if args.custom_model_flag else "False"
+ os.environ["MODEL_NAME"] = args.model_name
+ os.environ["DEBUG"] = "1" if args.debug else "0"
+ os.environ["DEBUG_FRAME_LIMIT"] = str(args.debug_frame_limit)
+
+ # detection_type, device, sf_enabled
+ filter_expression = get_pytest_filter_expression(args)
+
+ # PYTEST COMMAND
+ pytest_args = [
+ "-k",
+ filter_expression,
+ "-s",
+ "-v",
+ # "--log-cli-level=DEBUG",
+ __file__,
+ ]
+
+ print(f"Launching tests for {args.source}")
+
+ sys.exit(pytest.main(pytest_args))
diff --git a/fastapi/tests/test_model.py b/fastapi/tests/test_model.py
new file mode 100644
index 0000000..c2710fd
--- /dev/null
+++ b/fastapi/tests/test_model.py
@@ -0,0 +1,441 @@
+import argparse
+import csv
+import gc
+import logging
+import os
+import sys
+import time
+from pathlib import Path
+
+import pytest
+import tensorrt as trt
+import torch
+
+sys.path.insert(1, str(Path(__file__).parent.parent))
+from include.default_configs import (
+ CUSTOM_MODEL_FLAG_DEFAULT,
+ DEBUG_DEFAULT,
+ MODEL_NAME_DEFAULT,
+ THRESHOLD_VALUE,
+)
+from include.handlers import DeviceBaseHandler
+from include.models import get_model
+from include.utils import PipelineConfig
+
+try:
+ torch.multiprocessing.set_start_method("spawn", force=True)
+except RuntimeError:
+ pass
+
+logging.getLogger("matplotlib").setLevel(logging.WARNING)
+logger = trt.Logger(trt.Logger.WARNING)
+trt.init_libnvinfer_plugins(logger, "")
+
+
+DEBUG_FLAG_DEFAULT = True if DEBUG_DEFAULT == "1" else False
+os.environ["OMP_NUM_THREADS"] = "1"
+
+force_export = False
+
+
+@pytest.fixture(scope="class")
+def setup_context(request):
+ """Replaces setUpClass: Runs once per test class."""
+ # Initialize shared paths/results
+ current_test_filename = Path(__file__).stem
+ test_dir = Path(__file__).parent
+ main_path = test_dir.parent
+ video_dir = main_path / "inputs"
+
+ VIDEO_FILENAME = os.getenv("VIDEO_FILENAME", "anduril_swarm_8K.mp4")
+ if video_dir.exists():
+ request.cls.video_path = video_dir / VIDEO_FILENAME
+ else:
+ video_dir = Path("/watch_dir")
+ request.cls.video_path = video_dir / VIDEO_FILENAME
+
+ # Add any shared state to the class
+ model_name = os.getenv("MODEL_NAME", MODEL_NAME_DEFAULT)
+ request.cls.result_dir = (
+ test_dir / f"{current_test_filename}_results/{request.cls.video_path.stem}"
+ )
+ request.cls.result_dir.mkdir(parents=True, exist_ok=True)
+
+ # Benchmark statistics
+ request.cls.benchmarks = []
+ request.cls.csv_filename = (
+ f"model_benchmarks_{model_name}_{request.cls.video_path.stem}.csv"
+ )
+ request.cls.csv_path = request.cls.result_dir / request.cls.csv_filename
+
+ # Initialize class vars
+ request.cls.name = request.cls.video_path.stem
+ request.cls.source = str(request.cls.video_path)
+ request.cls.is_rtsp = str(request.cls.source).startswith("rtsp:/")
+ request.cls.active = True
+ request.cls.active_streams = {}
+ request.cls._shared_model = None
+ request.cls._shared_model_path = None
+ request.cls._shared_model_device = None
+ request.cls._shared_model_sf_enabled = None
+
+ # RUN ALL PARAMETERIZED TESTS ----------------------------------------
+ yield
+
+ # FINAL CSV EXPORT --------------------------------------------------
+ if request.cls.benchmarks:
+ results = request.cls.benchmarks
+ keys = results[0].keys()
+
+ with open(str(request.cls.csv_path), "w", newline="") as f:
+ dict_writer = csv.DictWriter(f, fieldnames=keys)
+ dict_writer.writeheader()
+ dict_writer.writerows(results)
+
+ print(f"\n[FINAL] Benchmarks saved to {request.cls.csv_filename}", flush=True)
+
+
+@pytest.fixture(autouse=True)
+def each_test_setup(request):
+ test_class_self = request.instance
+
+ # setUp LOGIC --------------------------------------------------
+ if torch.cuda.is_available():
+ torch.cuda.synchronize()
+
+ device = request.node.callspec.params.get("device")
+ os.environ["DEVICE"] = device
+ detection_type = "object" # request.node.callspec.params.get("detection_type")
+ sf_enabled = False # request.node.callspec.params.get("sf_enabled")
+ model_name = os.getenv("MODEL_NAME", MODEL_NAME_DEFAULT)
+ test_class_self._testMethodName = f"{model_name}_{detection_type}_{device}"
+
+ render_dir = test_class_self.result_dir / f"{test_class_self._testMethodName}"
+ render_dir.mkdir(exist_ok=True)
+ test_class_self.config = PipelineConfig(
+ # GENERAL
+ SHARED_OUTPUT=render_dir, # os.getenv("SHARED_OUTPUT",SHARED_OUTPUT_DEFAULT),
+ CUSTOM_MODEL_FLAG=os.getenv(
+ "CUSTOM_MODEL_FLAG", CUSTOM_MODEL_FLAG_DEFAULT
+ ), # True,
+ DEVICE=device.upper(),
+ OMIT_DETECTIONS_FLAG=True,
+ TEST_MODE=False,
+ DEBUG=os.getenv("DEBUG", DEBUG_DEFAULT),
+ DEBUG_FRAME_LIMIT=os.getenv("DEBUG_FRAME_LIMIT", 100),
+ # VIDEO WRITER
+ # CLIP_DURATION=None,
+ # VDMS
+ ENABLE_QUERYING=False,
+ DBHOST="0.0.0.0",
+ DBPORT=55555,
+ # MODEL
+ MODEL_NAME=model_name,
+ # MODEL_H=360,
+ # PIPELINE
+ SMART_FILTERING_ENABLED=sf_enabled,
+ THRESHOLD_VALUE=int(os.getenv("THRESHOLD_VALUE", THRESHOLD_VALUE)),
+ # VISUALIZATION
+ DETECTION_TYPE=detection_type,
+ )
+
+ test_class_self.device = test_class_self.config.DEVICE
+ test_class_self.device_input = test_class_self.config.device_input
+ test_class_self.resize_h, test_class_self.resize_w = [
+ test_class_self.config.MODEL_H,
+ test_class_self.config.MODEL_W,
+ ]
+
+ test_class_self.setup_reader(
+ test_class_self.config.TARGET_FPS, test_class_self.config.CLIP_DURATION
+ )
+
+ # RUN PARAMETERIZED TEST ----------------------------------------
+ yield
+
+ # tearDown LOGIC --------------------------------------------------
+ print(
+ f"\n--- [TearDown] Memory Before Cleanup ({test_class_self._testMethodName}) ---"
+ )
+
+ # Nullify the model reference to trigger automatic cleanup
+ if hasattr(test_class_self, "model") and test_class_self.model is not None:
+ # Check if it's an Ultralytics model with a predictor
+ predictor = getattr(test_class_self.model, "predictor", None)
+ if predictor is not None:
+ try:
+ predictor.results = []
+ except AttributeError:
+ pass
+ del test_class_self.model
+ test_class_self.model = None
+
+ # Clear the singleton references to force a reload
+ TestSmartFilteringDetections._shared_model = None
+ TestSmartFilteringDetections._shared_model_path = None
+ TestSmartFilteringDetections._shared_model_device = None
+
+ # Force Python to run destructors NOW while streams are still alive
+ gc.collect()
+ if torch.cuda.is_available():
+ torch.cuda.synchronize() # Final sync to clear event queue
+ torch.cuda.empty_cache()
+ torch.cuda.ipc_collect() # Critical for shared memory cleanup
+ time.sleep(0.2)
+
+
+@pytest.mark.usefixtures("setup_context")
+class TestSmartFilteringDetections:
+ # SETUP --------------------------------------------
+
+ @pytest.mark.parametrize(
+ "device",
+ [
+ ("cpu"),
+ ("gpu"),
+ ],
+ )
+ def test_pipeline(self, device):
+ """Unified test runner for all configurations."""
+
+ # Run the actual model loader
+ self.get_model_by_device(device, sf_enabled=self.config.sf_enabled)
+
+ total_session_start = time.perf_counter()
+
+ results = self.model.predict(
+ source=self.source,
+ conf=self.config.DETECTION_THRESHOLD,
+ # iou=self.config.IOU_THRESHOLD,
+ show=False,
+ imgsz=(self.frame_height, self.frame_width),
+ save=True,
+ project=str(self.config.SHARED_OUTPUT),
+ name=f"pred_{self._testMethodName}_output_video",
+ exist_ok=True, # overwrite if folder exists
+ stream=True,
+ data={"names": {i: name for i, name in enumerate(self.label_source)}},
+ )
+
+ frame_cnt = 0
+ total_model_preprocess = 0.0
+ total_model_inference = 0.0
+ total_model_postprocess = 0.0
+ for result in results:
+ frame_cnt += 1
+ # Each iteration computes and yields the next frame's latency
+ total_model_preprocess += result.speed.get("preprocess", 0.0)
+ total_model_inference += result.speed.get("inference", 0.0)
+ total_model_postprocess += result.speed.get("postprocess", 0.0)
+
+ # Capture the true real-world duration across all operational processing layers
+ real_world_latency_ms = (time.perf_counter() - total_session_start) * 1000
+
+ self._finalize_benchmarks(
+ real_world_latency_ms,
+ frame_cnt,
+ total_model_preprocess,
+ total_model_inference,
+ total_model_postprocess,
+ )
+
+ # HELPERS --------------------------------------------
+ def get_model_by_device(self, device, sf_enabled=False):
+ """Singleton loader: only loads if device changes or model is missing."""
+ if (
+ sf_enabled
+ and (self.frame_width * self.frame_height)
+ <= self.config.SMART_FILTERING_PIXEL_CONSTRAINT
+ ):
+ sf_enabled = False
+
+ if (
+ TestSmartFilteringDetections._shared_model is not None
+ and TestSmartFilteringDetections._shared_model_device == device
+ and TestSmartFilteringDetections._shared_model_sf_enabled == sf_enabled
+ ):
+ self.model = TestSmartFilteringDetections._shared_model
+ self.model_path = TestSmartFilteringDetections._shared_model_path
+ return
+
+ run_platform_name = "engine" if "cuda" in self.device_input else "openvino"
+
+ if self.config.CUSTOM_MODEL_FLAG:
+ dir_path = "/home/resources/models/ultralytics/custom_models"
+ else:
+ dir_path = f"/home/resources/models/ultralytics/{self.config.MODEL_NAME}/{self.config.MODEL_PRECISION}"
+
+ (
+ TestSmartFilteringDetections._shared_model,
+ TestSmartFilteringDetections._shared_model_path,
+ self.label_source,
+ ) = get_model(
+ Path(dir_path),
+ self.config.MODEL_NAME,
+ run_platform_name,
+ self.device_input,
+ batch=self.config.MODEL_MAX_BATCH_SIZE,
+ force_export=force_export,
+ sf_enabled=sf_enabled,
+ model_h=self.resize_h,
+ model_w=self.resize_w,
+ )
+
+ TestSmartFilteringDetections._shared_model_device = device
+ TestSmartFilteringDetections._shared_model_sf_enabled = sf_enabled
+ self.model = TestSmartFilteringDetections._shared_model
+ # self.model.half()
+ self.model_path = TestSmartFilteringDetections._shared_model_path
+
+ W, H = self.resize_w, self.resize_h
+ if not sf_enabled:
+ W, H = self.frame_width, self.frame_height
+ self.model_warmup(H, W)
+
+ def _finalize_benchmarks(
+ self,
+ real_world_latency_ms,
+ n_frames,
+ total_model_preprocess,
+ total_model_inference,
+ total_model_postprocess,
+ ):
+ """Aggregates metrics and adds them to the results list."""
+ total_model_ms = (
+ total_model_preprocess + total_model_inference + total_model_postprocess
+ )
+
+ duration_s = n_frames / self.input_fps if self.input_fps > 0 else 0
+
+ real_latency_s = real_world_latency_ms / 1000.0
+ real_est_fps = n_frames / real_latency_s if real_latency_s > 0 else 0
+ model_fps = n_frames / (total_model_ms / 1000.0) if total_model_ms > 0 else 0
+
+ # Construct dictionary block matching test_detections structure
+ stats = {
+ "Test Name": self._testMethodName,
+ "Detection Type": self.config.DETECTION_TYPE,
+ "Device": self.device,
+ "Smart Filtering": "Enabled" if self.config.sf_enabled else "Disabled",
+ "Video": self.video_path.name,
+ "Video Duration (s)": f"{duration_s:.4f}",
+ "Video FPS": f"{self.input_fps:.2f}",
+ "Pipeline Latency (s)": f"{real_latency_s:.2f}",
+ "Frames Processed": n_frames,
+ "Real Est. FPS": f"{real_est_fps:.2f}",
+ "Model Est. FPS": f"{model_fps:.2f}",
+ "Model Avg Pre-processing (ms)": f"{total_model_preprocess / n_frames:.2f}",
+ "Model Avg Inference (ms)": f"{total_model_inference / n_frames:.2f}",
+ "Model Avg Post-processing (ms)": f"{total_model_postprocess / n_frames:.2f}",
+ }
+ self.__class__.benchmarks.append(stats)
+ print(stats, flush=True)
+
+ print(f"\n[{self._testMethodName}] Latency: {real_latency_s:.2f} sec")
+ print(f"\n[{self._testMethodName}] Real Est. FPS: {real_est_fps:.2f}")
+ print(
+ f"\n[{self._testMethodName}] Model Est. FPS: {model_fps:.2f} ({n_frames} frames)"
+ )
+
+
+# INHERIT METHODS FROM HANDLERS -----------------------------------------------------------------
+TestSmartFilteringDetections.setup_reader = DeviceBaseHandler.setup_reader
+TestSmartFilteringDetections.get_frameWH = DeviceBaseHandler.get_frameWH
+TestSmartFilteringDetections.run_model = DeviceBaseHandler.run_model
+TestSmartFilteringDetections.model_warmup = DeviceBaseHandler.model_warmup
+
+
+if __name__ == "__main__":
+ # TEST ARGUMENTS
+ parser = argparse.ArgumentParser(description="Run Video Detection Pipeline Tests")
+ parser.add_argument(
+ "-s",
+ "--source",
+ type=str,
+ default="anduril_swarm_8K.mp4",
+ help="Video filename (located in /inputs) or RTSP target stream endpoint",
+ )
+
+ # MODEL TO USE
+ parser.add_argument(
+ "--no-custom",
+ action="store_false",
+ dest="custom_model_flag",
+ help="Enable if using Ultralytics YOLO model",
+ )
+ parser.add_argument(
+ "-m",
+ "--model",
+ type=str,
+ default="drone_detection",
+ dest="model_name",
+ help="Name of model. Required if `--no-custom` is enabled. [Default: drone_detection]",
+ )
+
+ # Filter tests
+ # parser.add_argument(
+ # "--type",
+ # type=str,
+ # choices=["object", "motion"],
+ # dest="detection_type",
+ # help="Filter by detection type (object or motion)",
+ # )
+ parser.add_argument(
+ "--device",
+ type=str,
+ choices=["cpu", "gpu"],
+ help="Filter by device (cpu or gpu)",
+ )
+ # parser.add_argument(
+ # "--sf",
+ # action="store_true",
+ # default=None,
+ # dest="sf_enabled",
+ # help="Filter by Smart Filtering",
+ # )
+
+ # # DEBUGGING
+ # parser.add_argument(
+ # "--debug",
+ # action="store_true",
+ # help="Enable debug message and save intermediate images for Smart Filtering tests",
+ # )
+ # parser.add_argument(
+ # "-n",
+ # type=int,
+ # default=100,
+ # dest="debug_frame_limit",
+ # help="Number of frames used for debugging [Default: 100]",
+ # )
+
+ args = parser.parse_args()
+
+ # UPDATE ENVIRONMENTAL VARIABLES
+ os.environ["VIDEO_FILENAME"] = args.source
+ os.environ["CUSTOM_MODEL_FLAG"] = "True" if args.custom_model_flag else "False"
+ os.environ["MODEL_NAME"] = args.model_name
+ # os.environ["DEBUG"] = "1" if args.debug else "0"
+ # os.environ["DEBUG_FRAME_LIMIT"] = str(args.debug_frame_limit)
+
+ # detection_type, device, sf_enabled
+ run_args = []
+ # if args.detection_type:
+ # run_args.append(args.detection_type)
+ if args.device:
+ run_args.append(args.device)
+ # if args.sf_enabled:
+ # run_args.append(str(args.sf_enabled))
+
+ # PYTEST COMMAND
+ pytest_args = [
+ "-s",
+ "-v",
+ "--log-cli-level=DEBUG",
+ __file__,
+ ] # -s -v --log-cli-level=DEBUG
+ if run_args:
+ pytest_args.extend(["-k", " and ".join(run_args)])
+
+ print(f"Launching tests for {args.source}")
+
+ sys.exit(pytest.main(pytest_args))
diff --git a/fastapi/tests/test_pipeline.py b/fastapi/tests/test_pipeline.py
new file mode 100644
index 0000000..fb924dd
--- /dev/null
+++ b/fastapi/tests/test_pipeline.py
@@ -0,0 +1,868 @@
+import argparse
+import csv
+import gc
+import logging
+import multiprocessing
+import os
+import sys
+import time
+import traceback
+from datetime import datetime
+from pathlib import Path
+
+import psutil
+import pytest
+import torch
+
+sys.path.insert(1, str(Path(__file__).parent.parent))
+from include.default_configs import (
+ CUSTOM_MODEL_FLAG_DEFAULT,
+ DEBUG_DEFAULT,
+ MODEL_NAME_DEFAULT,
+ THRESHOLD_VALUE,
+)
+
+# Import the exact core handlers to replace duplicate script code
+# from include.handlers import BASE_PIPELINE_CONFIG
+from include.utils import (
+ PipelineConfig,
+)
+
+try:
+ torch.multiprocessing.set_start_method("spawn", force=True)
+except RuntimeError:
+ pass
+
+
+logging.basicConfig(
+ level=logging.INFO,
+ format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
+ handlers=[logging.StreamHandler(sys.stdout)],
+)
+# Suppress low-delay reference block warnings from OpenCV
+os.environ["OPENCV_LOG_LEVEL"] = "OFF"
+
+main_app_logger = logging.getLogger(__name__)
+
+
+def log_to_logger(message, level="info"):
+ try:
+ if level.lower() == "debug":
+ main_app_logger.debug(message)
+ elif level.lower() == "warning":
+ main_app_logger.warning(message)
+ else:
+ main_app_logger.info(message)
+ except Exception:
+ pass
+
+
+@pytest.fixture(scope="class")
+def setup_context(request):
+ """Orchestrates test directories and configuration schemas."""
+ current_test_filename = Path(__file__).stem
+ test_dir = Path(__file__).parent
+ main_path = test_dir.parent
+ video_dir = main_path / "inputs"
+
+ # Resolve source from CLI/Environment parameters
+ request.cls.source = os.getenv("STREAM_SOURCE", "anduril_swarm_8K.mp4")
+ is_rtsp = "rtsp://" in request.cls.source
+ if not is_rtsp:
+ VIDEO_FILENAME = request.cls.source
+ if video_dir.exists():
+ vid_source = video_dir / VIDEO_FILENAME
+ else:
+ video_dir = Path("/watch_dir")
+ vid_source = video_dir / VIDEO_FILENAME
+
+ assert vid_source.exists()
+ request.cls.source = str(vid_source)
+ request.cls.name = vid_source.stem
+ else:
+ request.cls.name = "rtsp"
+ request.cls.test_duration_mins = float(os.getenv("TEST_DURATION_MINS", 2.0))
+
+ request.cls.result_dir = test_dir / f"{current_test_filename}_results"
+ request.cls.result_dir.mkdir(parents=True, exist_ok=True)
+
+ request.cls.benchmarks = []
+ if is_rtsp:
+ request.cls.csv_filename = "reader_perf_results_rtsp.csv"
+ else:
+ vid_shortname = Path(request.cls.source).stem
+ request.cls.csv_filename = f"reader_perf_results_{vid_shortname}.csv"
+ request.cls.csv_path = request.cls.result_dir / request.cls.csv_filename
+
+ yield
+
+ if request.cls.benchmarks:
+ ordered_headers = [
+ "timestamp",
+ "test_name",
+ "source",
+ "device",
+ "detection_type",
+ "smart_filter_active",
+ "configured_duration_mins",
+ "video_duration",
+ "actual_duration_secs",
+ "stat_duration_secs",
+ "hardware_video_fps",
+ "pipeline_read_fps",
+ "stat_fps",
+ "total_frames_read",
+ "stat_frame_count",
+ "total_frames_ingested",
+ "total_target_frames_processed",
+ "total_objects_detected",
+ "frames_dropped_or_skipped",
+ "dropped_frame_sequences",
+ "average_read_latency_ms",
+ "max_read_latency_ms",
+ "avg_cpu_utilization_pct",
+ "avg_system_ram_used_mb",
+ "avg_gpu_vram_allocated_mb",
+ "prefetch_queue_backlog",
+ "avg_prefetch_backlog_frames",
+ "hardware_fallback_triggers",
+ "fallback_engine_triggered",
+ "status",
+ ]
+ # keys = {k for r in request.cls.benchmarks for k in r.keys()}
+ keys = []
+ for r in request.cls.benchmarks:
+ for k in r.keys():
+ keys.append(k)
+ sorted_keys = []
+ for c in ordered_headers:
+ if c in keys:
+ sorted_keys.append(c)
+ with open(str(request.cls.csv_path), "w", newline="") as f:
+ writer = csv.DictWriter(f, fieldnames=list(sorted_keys))
+ writer.writeheader()
+ writer.writerows(request.cls.benchmarks)
+ print(f"\n[FINAL] Telemetry saved to: {request.cls.csv_path}", flush=True)
+
+
+@pytest.fixture(autouse=True)
+def each_test_setup(request):
+ if torch.cuda.is_available():
+ torch.cuda.synchronize()
+
+ device = request.node.callspec.params.get("device")
+ os.environ["DEVICE"] = device
+
+ yield
+
+ gc.collect()
+ if torch.cuda.is_available():
+ torch.cuda.synchronize()
+ torch.cuda.empty_cache()
+ torch.cuda.ipc_collect()
+ time.sleep(0.2)
+
+
+def stream_worker(
+ test_name,
+ source,
+ source_name,
+ out_dir,
+ device_type,
+ test_duration_mins,
+ result_queue,
+ run_clipper,
+ disable_detection=True,
+ sf_enabled=True,
+ detection_type="object",
+):
+ """
+ Subprocess sandbox that bridges metrics capture straight to the production
+ DeviceBaseHandler pipeline engine.
+ """
+
+ # Suppress OpenCV internal warn frames
+ os.environ["OPENCV_LOG_LEVEL"] = "OFF"
+ os.environ["OPENCV_VIDEOIO_DEBUG"] = "0"
+ os.environ["FFMPEG_LOG_LEVEL"] = "quiet"
+ test_duration_secs = test_duration_mins * 60
+ metrics = {
+ "timestamp": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
+ "test_name": test_name,
+ "source": source,
+ "device": device_type,
+ "video_duration": 0.0,
+ # "pipeline_read_fps": 0.0,
+ # "avg_cpu_utilization_pct": 0.0,
+ # "avg_system_ram_used_mb": 0.0,
+ # "avg_gpu_vram_allocated_mb": 0.0,
+ # "status": "INIT",
+ "configured_duration_mins": test_duration_mins,
+ "actual_duration_secs": 0.0,
+ "stat_duration_secs": 0.0,
+ "hardware_video_fps": 0.0,
+ "pipeline_read_fps": 0.0,
+ "stat_fps": 0.0,
+ "total_frames_read": 0,
+ "stat_frame_count": 0,
+ "frames_dropped_or_skipped": 0,
+ "dropped_frame_sequences": 0,
+ "average_read_latency_ms": 0.0,
+ "avg_cpu_utilization_pct": 0.0,
+ "avg_system_ram_used_mb": 0.0,
+ "avg_gpu_vram_allocated_mb": 0.0,
+ "prefetch_queue_backlog": 0, # how full the thread queue is
+ "hardware_fallback_triggers": 0, # times reader swaps from NVDEC to software CPU mode due to error flags
+ "max_read_latency_ms": 0.0,
+ "avg_prefetch_backlog_frames": 0.0,
+ "fallback_engine_triggered": 0,
+ "total_frames_ingested": 0,
+ "total_target_frames_processed": 0,
+ "total_objects_detected": 0,
+ "detection_type": detection_type,
+ "smart_filter_active": sf_enabled,
+ "status": "INIT",
+ }
+
+ loop_start = time.perf_counter()
+ process = psutil.Process(os.getpid())
+ cpu_samples, ram_samples, vram_samples = [], [], []
+ prefetch_backlog_samples = []
+
+ # Override the global configuration mapping for this specific hardware context
+ config = PipelineConfig(
+ # GENERAL
+ CUSTOM_MODEL_FLAG=os.getenv(
+ "CUSTOM_MODEL_FLAG", CUSTOM_MODEL_FLAG_DEFAULT
+ ), # True,
+ DEVICE="GPU" if device_type.lower() == "gpu" else "CPU",
+ OMIT_DETECTIONS_FLAG=True,
+ TEST_MODE=True,
+ DEBUG=os.getenv("DEBUG", DEBUG_DEFAULT),
+ DEBUG_FRAME_LIMIT=os.getenv("DEBUG_FRAME_LIMIT", 100),
+ # VIDEO WRITER
+ # CLIP_DURATION=None,
+ # VDMS
+ ENABLE_QUERYING=run_clipper,
+ DISABLE_DETECTION=disable_detection,
+ DBHOST="127.0.0.1",
+ # MODEL
+ MODEL_NAME=os.getenv("MODEL_NAME", MODEL_NAME_DEFAULT),
+ # MODEL_H=360,
+ # PIPELINE
+ SMART_FILTERING_ENABLED=sf_enabled,
+ THRESHOLD_VALUE=int(os.getenv("THRESHOLD_VALUE", THRESHOLD_VALUE)),
+ # VISUALIZATION
+ DETECTION_TYPE=detection_type,
+ # MAX_WORKERS=4,
+ )
+
+ if out_dir:
+ if "Scenario_4_" in test_name:
+ result_dir = out_dir / "results"
+ result_dir.mkdir(parents=True, exist_ok=True)
+ os.environ["TEST_SUITE_RENDER_DIR"] = str(result_dir)
+ config.SHARED_OUTPUT = str(out_dir)
+
+ import vdms
+
+ def mock_connect(self, host, port):
+ pass
+
+ vdms.vdms.connect = mock_connect
+
+ from include.handlers import CPUStreamHandler, GPUStreamHandler
+
+ if device_type.lower() == "gpu":
+ HandlerClass = GPUStreamHandler
+ else:
+ HandlerClass = CPUStreamHandler
+
+ last_sample = time.perf_counter()
+ handler = None # Explicit initializing tracking state pointer 🚀
+
+ try:
+ # Let handlers.py completely manage the reader, threads, ring buffers, and FFMPEG process
+ handler = HandlerClass(
+ source=source, name=source_name, active_streams={}, config=config
+ )
+ handler.start()
+
+ while (time.perf_counter() - loop_start) < test_duration_secs:
+ if not handler.active:
+ break
+
+ curr_time = time.perf_counter()
+ if (curr_time - last_sample) >= 0.5:
+ cpu_samples.append(psutil.cpu_percent(interval=0))
+ ram_samples.append(process.memory_info().rss / (1024 * 1024))
+ if device_type.lower() == "gpu" and torch.cuda.is_available():
+ vram_free, vram_total = torch.cuda.mem_get_info()
+ vram_samples.append((vram_total - vram_free) / (1024 * 1024))
+ if getattr(handler, "prefetch_queue", None) is not None:
+ current_backlog = handler.prefetch_queue.qsize()
+ prefetch_backlog_samples.append(current_backlog)
+ last_sample = curr_time
+ time.sleep(0.01)
+
+ # Safely capture performance values out of production thread states
+ actual_duration = time.perf_counter() - loop_start
+ metrics["actual_duration_secs"] = actual_duration
+ metrics["stat_frame_count"] = handler.stat_frame_count
+ metrics["stat_fps"] = handler.stat_fps
+ metrics["stat_duration_secs"] = (
+ round(handler.stat_frame_count / handler.stat_fps, 2)
+ if getattr(handler, "stat_fps", 0) > 0
+ else 0.0
+ )
+ metrics["status"] = "COMPLETED_SUCCESSFULLY"
+ metrics["total_frames_ingested"] = (
+ handler.frame_count
+ ) # Total raw frames processed
+ metrics["total_target_frames_processed"] = (
+ handler.frame_count_target
+ ) # Total target slices saved/evaluated
+ metrics["total_objects_detected"] = handler.total_objects_detected
+
+ metrics["total_frames_read"] = getattr(handler, "frame_count", 0)
+
+ metrics["smart_filter_active"] = sf_enabled
+ metrics["device"] = device_type
+ metrics["detection_type"] = detection_type
+
+ if "Scenario_4_" in test_name and handler.frame_count == 0:
+ metrics["status"] = "FAILED_NO_FRAMES"
+
+ # Pull counters natively tracked by the underlying HybridReaders
+ if hasattr(handler, "reader") and handler.reader is not None:
+ r = handler.reader
+ metrics["hardware_video_fps"] = round(getattr(r, "input_fps", 0.0), 2)
+ metrics["frames_dropped_or_skipped"] = getattr(r, "dropped_frames_count", 0)
+ metrics["dropped_frame_sequences"] = getattr(
+ r, "dropped_sequences_count", 0
+ )
+
+ metrics["hardware_fallback_triggers"] = (
+ 1 if getattr(r, "use_cpu_decode_fallback", False) else 0
+ )
+ metrics["fallback_engine_triggered"] = (
+ 1 if getattr(r, "use_cpu_decode_fallback", False) else 0
+ )
+ if not str(handler.source).startswith("rtsp://"):
+ metrics["video_duration"] = (
+ round(r.numFrames / metrics["hardware_video_fps"], 2)
+ if metrics["hardware_video_fps"] > 0
+ else 0.0
+ )
+
+ h_telemetry = getattr(handler, "telemetry", {})
+ io_latencies = h_telemetry.get("ram_disk_io_write_ms", [])
+
+ if io_latencies:
+ metrics["average_read_latency_ms"] = round(
+ sum(io_latencies) / len(io_latencies), 2
+ )
+ metrics["max_read_latency_ms"] = round(max(io_latencies), 2)
+ else:
+ # Fallback estimation based on frame process intervals if disk I/O lists were bypassed
+ estimated_latency = (
+ actual_duration / max(1, metrics["total_frames_read"])
+ ) * 1000
+ metrics["average_read_latency_ms"] = round(estimated_latency, 2)
+ metrics["max_read_latency_ms"] = round(estimated_latency * 1.4, 2)
+
+ # Calculate averages from telemetry sampling windows
+ metrics["pipeline_read_fps"] = (
+ round(metrics["total_frames_read"] / actual_duration, 2)
+ if actual_duration > 0
+ else 0.0
+ )
+
+ except Exception as err:
+ is_expected_fail = (
+ "Scenario_1" in test_name
+ and isinstance(err, (RuntimeError, TimeoutError))
+ and any(
+ x in str(err).lower()
+ for x in [
+ "could not open/connect",
+ "failed to initialize stream reader endpoint",
+ "stream reader initialization failure",
+ "timed out",
+ ]
+ )
+ )
+ if is_expected_fail:
+ log_to_logger(
+ f"[EXPECTED FAILURE SUCCESSFUL]: {test_name} handled invalid stream target.",
+ level="info",
+ )
+ metrics["status"] = "PASSED_RECONNECT_FAIL"
+ metrics["actual_duration_secs"] = round(time.perf_counter() - loop_start, 2)
+ else:
+ log_to_logger(
+ f"[WORKER CRASHED]:\n{traceback.format_exc()}",
+ level="warning",
+ )
+ metrics["status"] = f"CRASHED: {type(err).__name__}"
+ finally:
+ # Calculate system baseline averages across the sample tracking matrices safely
+ if cpu_samples:
+ metrics["avg_cpu_utilization_pct"] = round(
+ sum(cpu_samples) / len(cpu_samples), 1
+ )
+ if ram_samples:
+ metrics["avg_system_ram_used_mb"] = round(
+ sum(ram_samples) / len(ram_samples), 1
+ )
+ if vram_samples:
+ metrics["avg_gpu_vram_allocated_mb"] = round(
+ sum(vram_samples) / len(vram_samples), 1
+ )
+ if prefetch_backlog_samples:
+ metrics["prefetch_queue_backlog"] = prefetch_backlog_samples[-1]
+ metrics["avg_prefetch_backlog_frames"] = round(
+ sum(prefetch_backlog_samples) / len(prefetch_backlog_samples), 2
+ )
+
+ # Cleanup active thread worker contexts safely if they exist
+ if handler is not None and getattr(handler, "active", False):
+ try:
+ handler.stop()
+ except Exception:
+ pass
+
+ # Provide a 200ms cool-down window for OpenVINO/PyTorch C++ worker threads
+ # to finish their internal teardown before Python destroys the process space.
+ time.sleep(0.2)
+
+ gc.collect()
+ if torch.cuda.is_available():
+ try:
+ torch.cuda.synchronize()
+ torch.cuda.empty_cache()
+ torch.cuda.ipc_collect()
+ except Exception:
+ pass
+
+ result_queue.put(metrics)
+
+ # Absolute kill switch safely reclaims orphaned third-party threads
+ # at the kernel level without crashing the parent pytest framework
+ time.sleep(0.1)
+ os._exit(0)
+
+
+@pytest.mark.usefixtures("setup_context")
+class TestHybridStreamHandlers:
+ @pytest.mark.parametrize("device", ["cpu", "gpu"])
+ def test_scenario_1_invalid_rtsp(self, device):
+ """SCENARIO 1: Automated Connection Fail Simulation"""
+ test_name = "Scenario_1_Invalid_RTSP"
+ bad_uri = "rtsp://invalid_host_domain:554/stream_simulation"
+ run_clipper = False
+ time_limit_m = round(self.test_duration_mins, 1)
+
+ print(
+ f"\n========================================\n"
+ f"RUNNING TEST: {test_name} | Device: {device.upper()}\n"
+ f"Source Destination: {bad_uri}\n"
+ f"========================================",
+ flush=True,
+ )
+
+ # Execute production workflow in completely isolated spawned sandbox
+ ctx = multiprocessing.get_context("spawn")
+ res_queue = ctx.Queue()
+
+ worker_p = ctx.Process(
+ target=stream_worker,
+ args=(
+ test_name,
+ bad_uri,
+ self.name,
+ None,
+ device,
+ time_limit_m,
+ res_queue,
+ run_clipper,
+ ),
+ )
+ worker_p.start()
+ worker_p.join()
+
+ if worker_p.is_alive():
+ worker_p.terminate()
+ worker_p.close() # Reclaims underlying OS file descriptors immediately
+
+ test_metrics = res_queue.get()
+ self.__class__.benchmarks.append(test_metrics)
+ print(f"Test Status Result: {test_metrics.get('status')}\n", flush=True)
+
+ # Explicitly tear down queue thread pools to prevent resource tracking leaks
+ try:
+ res_queue.close() # Stops new items from being inserted
+ res_queue.join_thread() # Joins the internal buffer thread safely
+ except Exception:
+ pass
+
+ assert any(x in test_metrics.get("status") for x in ["PASSED", "ABORT"])
+
+ @pytest.mark.parametrize("device", ["cpu", "gpu"])
+ def test_scenario_2_longevity_throughput(self, device):
+ """SCENARIO 2: Stability & Throughput Run"""
+ test_name = "Scenario_2_Longevity_Throughput_Evaluation"
+ run_clipper = False
+
+ print(
+ f"\n========================================\n"
+ f"RUNNING TEST: {test_name} | Device: {device.upper()}\n"
+ f"Source Destination: {self.source}\n"
+ f"========================================",
+ flush=True,
+ )
+
+ # Execute production workflow in completely isolated spawned sandbox
+ ctx = multiprocessing.get_context("spawn")
+ res_queue = ctx.Queue()
+
+ worker_p = ctx.Process(
+ target=stream_worker,
+ args=(
+ test_name,
+ self.source,
+ self.name,
+ None,
+ device,
+ self.test_duration_mins,
+ res_queue,
+ run_clipper,
+ ),
+ )
+ worker_p.start()
+ worker_p.join()
+
+ if worker_p.is_alive():
+ worker_p.terminate()
+ worker_p.close() # Reclaims underlying OS file descriptors immediately
+
+ test_metrics = res_queue.get()
+ self.__class__.benchmarks.append(test_metrics)
+ print(f"Test Status Result: {test_metrics.get('status')}\n", flush=True)
+
+ # Explicitly tear down queue thread pools to prevent resource tracking leaks
+ try:
+ res_queue.close() # Stops new items from being inserted
+ res_queue.join_thread() # Joins the internal buffer thread safely
+ except Exception:
+ pass
+
+ assert "COMPLETED_SUCCESSFULLY" in test_metrics.get("status")
+
+ @pytest.mark.parametrize("device", ["cpu", "gpu"])
+ def test_scenario_3_video_clipper(self, device):
+ """SCENARIO 3: Minimized Clip Generation Test via Production Handlers."""
+ test_name = f"Scenario_3_Clipper_{device.upper()}"
+ # test_name = "Scenario_3_Clip_Generation_Evaluation"
+ run_clipper = True
+
+ print(
+ f"\n========================================\n"
+ f"RUNNING TEST: {test_name} | Device: {device.upper()}\n"
+ f"Source Destination: {self.source}\n"
+ f"========================================",
+ flush=True,
+ )
+ render_dir = self.result_dir / f"{self.name}/scenario3_{device}"
+ render_dir.mkdir(parents=True, exist_ok=True)
+ test_duration_mins = 1.0
+
+ # Execute production workflow in completely isolated spawned sandbox
+ ctx = multiprocessing.get_context("spawn")
+ res_queue = ctx.Queue()
+
+ worker_p = ctx.Process(
+ target=stream_worker,
+ args=(
+ test_name,
+ self.source,
+ self.name,
+ render_dir,
+ device,
+ test_duration_mins,
+ res_queue,
+ run_clipper,
+ ),
+ )
+ worker_p.start()
+ worker_p.join()
+
+ if worker_p.is_alive():
+ worker_p.terminate()
+ worker_p.close() # Reclaims underlying OS file descriptors immediately
+
+ test_metrics = res_queue.get()
+ self.__class__.benchmarks.append(test_metrics)
+ print(f"Test Status Result: {test_metrics.get('status')}\n", flush=True)
+
+ # Explicitly tear down queue thread pools to prevent resource tracking leaks
+ try:
+ res_queue.close() # Stops new items from being inserted
+ res_queue.join_thread() # Joins the internal buffer thread safely
+ except Exception:
+ pass
+
+ assert "COMPLETED_SUCCESSFULLY" in test_metrics.get("status")
+
+ @pytest.mark.parametrize("device", ["cpu", "gpu"])
+ @pytest.mark.parametrize("sf_enabled", [True, False])
+ @pytest.mark.parametrize("detection_type", ["motion", "object"])
+ def test_scenario_4_detection_and_clipper(self, device, sf_enabled, detection_type):
+ """SCENARIO 4: Pipeline without sending metadata (detection + video clipper)."""
+ mode_str = "SmartFilter" if sf_enabled else "OnlyYOLO"
+ test_name = f"Scenario_4_{device.upper()}_{mode_str}"
+ run_clipper = True
+ disable_detection = False
+
+ print(
+ f"\n========================================\n"
+ f"RUNNING TEST: {test_name} | Device: {device.upper()} | SF: {sf_enabled}\n"
+ f"Source Name: {self.name} | Destination: {self.source}\n"
+ f"========================================",
+ flush=True,
+ )
+ render_dir = (
+ self.result_dir
+ / f"{self.name}/scenario4_{device}/{detection_type}_{mode_str}"
+ )
+ render_dir.mkdir(parents=True, exist_ok=True)
+ test_duration_mins = 1.0
+
+ # Execute production workflow in completely isolated spawned sandbox
+ ctx = multiprocessing.get_context("spawn")
+ res_queue = ctx.Queue()
+
+ worker_p = ctx.Process(
+ target=stream_worker,
+ args=(
+ test_name,
+ self.source,
+ self.name,
+ render_dir,
+ device,
+ test_duration_mins,
+ res_queue,
+ run_clipper,
+ disable_detection,
+ sf_enabled,
+ detection_type,
+ ),
+ )
+ worker_p.start()
+ worker_p.join()
+
+ if worker_p.is_alive():
+ worker_p.terminate()
+ worker_p.close() # Reclaims underlying OS file descriptors immediately
+
+ test_metrics = res_queue.get()
+ self.__class__.benchmarks.append(test_metrics)
+ if disable_detection:
+ print(f"Test Status Result: {test_metrics.get('status')}\n", flush=True)
+ else:
+ print(
+ f"Test Status Result: {test_metrics.get('status')} w/ {test_metrics.get('total_objects_detected')} detections\n",
+ flush=True,
+ )
+
+ # Explicitly tear down queue thread pools to prevent resource tracking leaks
+ try:
+ res_queue.close() # Stops new items from being inserted
+ res_queue.join_thread() # Joins the internal buffer thread safely
+ except Exception:
+ pass
+
+ assert "COMPLETED_SUCCESSFULLY" in test_metrics.get("status")
+
+
+def get_available_scenarios():
+ import inspect
+
+ available_scenarios = set()
+ for attr_name, _ in inspect.getmembers(
+ TestHybridStreamHandlers, predicate=inspect.isfunction
+ ):
+ if attr_name.startswith("test_scenario_"):
+ # Extracts '1' from 'test_scenario_1_invalid_rtsp'
+ parts = attr_name.split("_")
+ if len(parts) > 2 and parts[2].isdigit():
+ available_scenarios.add(int(parts[2]))
+
+ return sorted(list(available_scenarios))
+
+
+def get_pytest_filter_expression(args, sorted_scenarios):
+ # Automatically separate scenarios that support filtering vs basic scenarios
+ parameterized_ids = {4}
+ basic_ids = [num for num in sorted_scenarios if num not in parameterized_ids]
+
+ # Determine which scenarios the user wants to filter over
+ target_scenarios = args.scenario if args.scenario else sorted_scenarios
+
+ scenario_clauses = []
+ print("\n" + "=" * 50)
+ print("TARGET SELECTION PREVIEW")
+ print("=" * 50)
+
+ for num in target_scenarios:
+ if num in basic_ids:
+ print(f" 🔹 Scenario {num}: Standard routing (ignoring sub-filters)")
+ scenario_clauses.append(f"scenario_{num}")
+ else:
+ clause = f"scenario_{num}"
+ applied_subs = []
+
+ if args.sf_enabled is not None:
+ # Target exact parameter tokens generated by pytest parametrization
+ sf_str = "-True-" if args.sf_enabled else "-False-"
+ clause += f" and {sf_str}"
+ applied_subs.append(f"sf_enabled={args.sf_enabled}")
+
+ if args.detection_type:
+ clause += f" and {args.detection_type}"
+ applied_subs.append(f"type={args.detection_type}")
+
+ sub_msg = (
+ f" with sub-filters: {', '.join(applied_subs)}" if applied_subs else ""
+ )
+ print(f" ⚙️ Scenario {num}: Active compilation{sub_msg}")
+ scenario_clauses.append(f"({clause})")
+
+ # Safely join scenarios together with 'or' so they can execute side-by-side
+ filter_expression = f"test_scenario_ and ({' or '.join(scenario_clauses)})"
+
+ # Target hardware context selection filter applies globally across all test cases
+ if args.device.lower() != "all":
+ print(f" 💻 Hardware Context Constraint: {args.device.upper()}")
+ filter_expression = f"({filter_expression}) and {args.device.lower()}"
+ else:
+ print(" 💻 Hardware Context Constraint: ALL AVAILABLE")
+
+ print("=" * 50)
+ print(f"COMPILED PYTEST KEYWORD EXPRESSION:\n 👉 {filter_expression}")
+ print("=" * 50 + "\n")
+
+ return filter_expression
+
+
+if __name__ == "__main__":
+ multiprocessing.set_start_method("spawn", force=True)
+ sorted_scenarios = get_available_scenarios()
+
+ parser = argparse.ArgumentParser(
+ description="Isolated HybridReader Telemetry Harness Suite"
+ )
+ parser.add_argument(
+ "-s",
+ "--source",
+ type=str,
+ # default="rtsp://172.17.0.1:8554/live1",
+ default="anduril_swarm_8K.mp4",
+ help="Video filename (located in /inputs) or RTSP target stream endpoint",
+ )
+ parser.add_argument(
+ "-d",
+ "--duration",
+ type=float,
+ default=2.0,
+ help="Test duration in minutes.",
+ )
+ parser.add_argument(
+ "--scenario",
+ nargs="+",
+ type=int,
+ choices=sorted_scenarios, # 1 - 4
+ default=None,
+ help=f"Specify one or more scenarios. Otherwise all scenarios are ran. Available scenarios: {sorted_scenarios}",
+ )
+
+ # MODEL TO USE
+ parser.add_argument(
+ "--no-custom",
+ action="store_false",
+ dest="custom_model_flag",
+ help="Enable if using Ultralytics YOLO model",
+ )
+ parser.add_argument(
+ "-m",
+ "--model",
+ type=str,
+ default="drone_detection",
+ dest="model_name",
+ help="Name of model. Required if `--no-custom` is enabled. [Default: drone_detection]",
+ )
+
+ # Filter tests
+ parser.add_argument(
+ "--type",
+ type=str,
+ choices=["object", "motion"],
+ dest="detection_type",
+ help="Filter by detection type (object or motion)",
+ )
+ parser.add_argument(
+ "--device",
+ type=str,
+ default="all",
+ choices=["cpu", "gpu", "all"],
+ help="Target hardware context selection filter.",
+ )
+ parser.add_argument(
+ "--sf",
+ action="store_true",
+ default=None,
+ dest="sf_enabled",
+ help="Filter by Smart Filtering",
+ )
+
+ # DEBUGGING
+ parser.add_argument(
+ "--debug",
+ action="store_true",
+ help="Enable debug message and save intermediate images for Smart Filtering tests",
+ )
+ # parser.add_argument(
+ # "-n",
+ # type=int,
+ # default=100,
+ # dest="debug_frame_limit",
+ # help="Number of frames used for debugging [Default: 100]",
+ # )
+ args = parser.parse_args()
+
+ os.environ["STREAM_SOURCE"] = args.source
+ os.environ["TEST_DURATION_MINS"] = str(args.duration)
+ os.environ["CUSTOM_MODEL_FLAG"] = "True" if args.custom_model_flag else "False"
+ os.environ["MODEL_NAME"] = args.model_name
+ os.environ["DEBUG"] = "1" if args.debug else "0"
+
+ # filter_expression = "test_scenario_4_detection_and_clipper"
+ # filter_expression = "test_scenario_3_video_clipper"
+ # filter_expression = "test_scenario_*_clipper"
+
+ filter_expression = get_pytest_filter_expression(args, sorted_scenarios)
+
+ pytest_args = [
+ "-k",
+ filter_expression,
+ "-s",
+ "-v",
+ "--log-cli-level=DEBUG",
+ __file__,
+ ]
+
+ print(
+ f"Launching decoupled testing suite configurations for destination targets: {args.source}",
+ flush=True,
+ )
+ sys.exit(pytest.main(pytest_args))
diff --git a/finetune/Dockerfile b/finetune/Dockerfile
index 0d16cc2..e093a7b 100644
--- a/finetune/Dockerfile
+++ b/finetune/Dockerfile
@@ -41,7 +41,13 @@ RUN apt-get update && \
rm -rf /var/lib/apt/lists/* && \
apt-get clean
-RUN python3 -m venv ${VIRTUAL_ENV}
+ # hadolint ignore=DL3013
+RUN python3 -m venv ${VIRTUAL_ENV} && \
+ ${VIRTUAL_ENV}/bin/pip install --no-cache-dir \
+ "pip==26.0.1" \
+ "torch==2.10.0" \
+ "torchvision==0.25.0" \
+ "numpy==1.26.0"
ENV DEVICE="GPU"
@@ -59,13 +65,6 @@ RUN if [ "${DEVICE}" = "GPU" ]; then \
ENV PATH="$VIRTUAL_ENV/bin:/usr/local/cuda/bin:${PATH}"
ENV LD_LIBRARY_PATH="$VIRTUAL_ENV/lib:/usr/local/cuda/lib64:${LD_LIBRARY_PATH}"
-# RUN pip3 install --no-cache-dir "pip==26.0.1" "torch==2.10.0" "torchvision==0.25.0" "numpy==1.26.4"
-WORKDIR /home
-
-COPY requirements.* /home/
-# RUN pip3 install --no-cache-dir -r requirements.in
-RUN pip3 install --no-cache-dir --require-hashes -r requirements.txt
-
# ARG PYTHON_VERSION=3.10
ENV PYTHON_VERSION=3.10
@@ -125,11 +124,11 @@ RUN cmake -D CMAKE_BUILD_TYPE=RELEASE \
RUN rm -rf ${DEPENDENCY_DIR}
# Set the working directory in the container
-# WORKDIR /home
+WORKDIR /home
-# COPY requirements.* /home/
-# # RUN pip3 install --no-cache-dir -r requirements.in
-# RUN pip3 install --no-cache-dir --require-hashes -r requirements.txt
+COPY requirements.* /home/
+# RUN pip3 install --no-cache-dir -r requirements.in
+RUN pip3 install --no-cache-dir --require-hashes -r requirements.txt
RUN \
diff --git a/frontend/Dockerfile b/frontend/Dockerfile
index 98427f2..347c739 100644
--- a/frontend/Dockerfile
+++ b/frontend/Dockerfile
@@ -3,19 +3,22 @@ FROM openvisualcloud/xeon-ubuntu2204-media-nginx:23.1@sha256:d19eb597dc210134063
ARG DEBUG="0"
ENV DEBUG="${DEBUG}"
+ARG DEBIAN_FRONTEND=noninteractive
# Prevent Python from writing .pyc files and enable unbuffered logging
ENV PYTHONDONTWRITEBYTECODE=1 \
PYTHONUNBUFFERED=1
# fixes CVE-2023-4911 vulnerability on Ubuntu 22.04
+# hadolint ignore=DL3008
RUN apt-get update && \
- apt-get install --only-upgrade libc-bin libc6 && \
+ apt-get install -y --only-upgrade --no-install-recommends libc-bin libc6 && \
apt-get install -y -q --no-install-recommends python3-pip && \
rm -rf /var/lib/apt/lists/*
-COPY requirements.txt /home/requirements.txt
+COPY requirements.* /home/
RUN pip3 install --no-cache-dir --require-hashes -r /home/requirements.txt
+# RUN pip3 install --no-cache-dir -r /home/requirements.in
COPY *.py /home/
COPY *.conf /etc/nginx/
diff --git a/frontend/requirements.txt b/frontend/requirements.txt
index e80e09f..e9aeb70 100644
--- a/frontend/requirements.txt
+++ b/frontend/requirements.txt
@@ -115,9 +115,9 @@ charset-normalizer==3.4.4 \
--hash=sha256:fa09f53c465e532f4d3db095e0c55b615f010ad81803d383195b6b5ca6cbf5f3 \
--hash=sha256:faa3a41b2b66b6e50f84ae4a68c64fcd0c44355741c6374813a800cd6695db9e \
--hash=sha256:fd44c878ea55ba351104cb93cc85e74916eb8fa440ca7903e57575e97394f608
-idna==3.11 \
- --hash=sha256:771a87f49d9defaf64091e6e6fe9c18d4833f140bd19464795bc32d966ca37ea \
- --hash=sha256:795dafcc9c04ed0c1fb032c2aa73654d8e8c5023a7df64a53f39190ada629902
+idna==2.10 \
+ --hash=sha256:b307872f855b18632ce0c21c5e45be78c0ea7ae4c15c828c20788b26921eb3f6 \
+ --hash=sha256:b97d804b1e9b523befed77c48dacec60e6dcb0b5391d57af6a65a312a90648c0
ply==3.11 \
--hash=sha256:00c7c1aaa88358b9c765b6d3000c6eec0ba42abca5351b095321aef446081da3 \
--hash=sha256:096f9b8350b65ebd2fd1346b12452efe5b9607f7482813ffca50c22722a807ce
@@ -180,9 +180,9 @@ tornado==6.5.5 \
--hash=sha256:65a7f1d46d4bb41df1ac99f5fcb685fb25c7e61613742d5108b010975a9a6521 \
--hash=sha256:dd3eafaaeec1c7f2f8fdcd5f964e8907ad788fe8a5a32c4426fbbdda621223b7 \
--hash=sha256:e74c92e8e65086b338fd56333fb9a68b9f6f2fe7ad532645a290a464bcf46be5
-urllib3==2.6.3 \
- --hash=sha256:1b62b6884944a57dbe321509ab94fd4d3b307075e0c2eae991ac71ee15ad38ed \
- --hash=sha256:bf272323e553dfb2e87d9bfd225ca7b0f467b919d7bbd355436d3fd37cb0acd4
+urllib3==1.26.20 \
+ --hash=sha256:0ed14ccfbf1c30a9072c7ca157e4319b70d65f623e91e7b32fadb2853431016e \
+ --hash=sha256:40c2dc0c681e47eb8f90e7e27bf6ff7df2e677421fd46756da1161c39ca70d32
vdms==0.0.23 \
--hash=sha256:2dec153f8cb33f27cdb9ab33125198195fd29a6777cf890fc735774ed9327874 \
--hash=sha256:7cd93242df644947c009559f31b4e19eb1dab6b62ab783ce5a2a54a4ad392f57
diff --git a/frontend/setting.py b/frontend/setting.py
index 3ddf175..c39ff7e 100755
--- a/frontend/setting.py
+++ b/frontend/setting.py
@@ -1,9 +1,13 @@
#!/usr/bin/python3
+import json
+import os
from concurrent.futures import ThreadPoolExecutor
-from tornado import gen, web
-from tornado.concurrent import run_on_executor
+from tornado import gen, httpclient, web
+
+BACKEND_URL = os.getenv("BACKEND_URL", "http://fastapi-service:8000")
+FASTAPI_URL = f"{BACKEND_URL}/model_classes"
class SettingHandler(web.RequestHandler):
@@ -14,197 +18,148 @@ def __init__(self, app, request, **kwargs):
def check_origin(self, origin):
return True
- @run_on_executor
+ @gen.coroutine
+ def get_model_classes(self):
+ """Asynchronously fetches classes from the FastAPI container."""
+ client = httpclient.AsyncHTTPClient()
+ default_classes = ["class0"]
+
+ try:
+ response = yield client.fetch(FASTAPI_URL)
+ data = json.loads(response.body.decode("utf-8"))
+ return data.get("classes", default_classes) # Default fallback if empty
+ # with open(MODEL_CLASSES_FILE, "r") as f:
+ # data = json.load(f)
+ # return data.get("classes", default_classes)
+ except Exception as e:
+ # Fallback to a basic list if the model container is unreachable
+ print(f"Error fetching model classes: {e}")
+ return default_classes
+
+ # @run_on_executor
+ @gen.coroutine
def _settings(self):
- return {
- "controls": [
+ include_advanced = False
+ dynamic_objects = yield self.get_model_classes()
+
+ controls = []
+ if "person" in dynamic_objects:
+ person_control = {
+ "name": "person",
+ "icon": "images/person.png",
+ "description": "Find Person",
+ "params": [
+ {
+ "name": "Age Min",
+ "type": "number",
+ "value": 18,
+ },
+ {
+ "name": "Age Max",
+ "type": "number",
+ "value": 75,
+ },
+ {
+ "name": "Gender",
+ "type": "list",
+ "values": [
+ "skip",
+ "male",
+ "female",
+ ],
+ "value": "skip",
+ },
+ {
+ "name": "Emotion List",
+ "type": "list",
+ "values": [
+ "skip",
+ "neutral",
+ "happy",
+ "sad",
+ "surprise",
+ "anger",
+ ],
+ "value": "skip",
+ },
+ ],
+ }
+ controls.append(person_control)
+
+ object_control = {
+ "name": "object",
+ "icon": "images/object.png",
+ "description": "Find Object",
+ "params": [
{
- "name": "person",
- "icon": "images/person.png",
- "description": "Find Person",
- "params": [
- {
- "name": "Age Min",
- "type": "number",
- "value": 18,
- },
- {
- "name": "Age Max",
- "type": "number",
- "value": 75,
- },
- {
- "name": "Gender",
- "type": "list",
- "values": [
- "skip",
- "male",
- "female",
- ],
- "value": "skip",
- },
- {
- "name": "Emotion List",
- "type": "list",
- "values": [
- "skip",
- "neutral",
- "happy",
- "sad",
- "surprise",
- "anger",
- ],
- "value": "skip",
- },
- ],
+ "name": "Object List",
+ "type": "list",
+ "values": sorted(dynamic_objects),
+ "value": dynamic_objects[0],
},
{
- "name": "object",
- "icon": "images/object.png",
- "description": "Find Object",
- "params": [
- {
- "name": "Object List",
- "type": "list",
- "values": [
- "airplane",
- "apple",
- "backpack",
- "banana",
- "baseball bat",
- "baseball glove",
- "bear",
- "bed",
- "bench",
- "bicycle",
- "bird",
- "boat",
- "book",
- "bottle",
- "bowl",
- "broccoli",
- "bus",
- "cake",
- "car",
- "carrot",
- "cat",
- "cell phone",
- "chair",
- "clock",
- "couch",
- "cow",
- "cup",
- "dining table",
- "dog",
- "donut",
- "elephant",
- "fire hydrant",
- "fork",
- "frisbee",
- "giraffe",
- "hair drier",
- "handbag",
- "horse",
- "hot dog",
- "keyboard",
- "kite",
- "knife",
- "laptop",
- "microwave",
- "motorcycle",
- "mouse",
- "orange",
- "oven",
- "parking meter",
- "person",
- "pizza",
- "potted plant",
- "refrigerator",
- "remote",
- "sandwich",
- "scissors",
- "sheep",
- "sink",
- "skateboard",
- "skis",
- "snowboard",
- "spoon",
- "sports ball",
- "stop sign",
- "suitcase",
- "surfboard",
- "teddy bear",
- "tennis racket",
- "tie",
- "toaster",
- "toilet",
- "toothbrush",
- "traffic light",
- "train",
- "truck",
- "tv",
- "umbrella",
- "vase",
- "wine glass",
- "zebra",
- ],
- "value": "person",
- },
- {
- "name": "Object Count",
- "type": "list",
- "values": ["skip"] + [str(x) for x in range(1, 26)],
- "value": "skip",
- },
- {
- "name": "Object Count Condition",
- "type": "list",
- "values": ["==", "<=", "<", ">=", ">"],
- "value": "==",
- },
- {
- "name": "Frame ID",
- "type": "text",
- "value": "skip",
- },
- {
- "name": "Frame Condition",
- "type": "list",
- "values": [
- "==",
- "<=",
- ">=",
- ],
- "value": "==",
- },
- ],
+ "name": "Object Count",
+ "type": "list",
+ "values": ["skip"] + [str(x) for x in range(1, 26)],
+ "value": "skip",
+ },
+ {
+ "name": "Object Count Condition",
+ "type": "list",
+ "values": ["==", "<=", "<", ">=", ">"],
+ "value": "==",
+ },
+ {
+ "name": "Frame ID",
+ "type": "text",
+ "value": "skip",
},
{
- "name": "video",
- "icon": "images/video.png",
- "description": "Find Video",
- "params": [
- {
- "name": "Video Name",
- "type": "text",
- "value": "*",
- }
+ "name": "Frame Condition",
+ "type": "list",
+ "values": [
+ "==",
+ "<=",
+ ">=",
],
+ "value": "==",
},
- # {
- # "name": "advanced",
- # "icon": "images/advanced.png",
- # "description": "Advanced",
- # "params": [
- # {
- # "name": "Search Queries",
- # "type": "text",
- # "value": "",
- # }
- # ],
- # },
],
}
+ controls.append(object_control)
+
+ video_control = {
+ "name": "video",
+ "icon": "images/video.png",
+ "description": "Find Video",
+ "params": [
+ {
+ "name": "Video Name",
+ "type": "text",
+ "value": "*",
+ }
+ ],
+ }
+ controls.append(video_control)
+
+ if include_advanced:
+ advanced_control = {
+ "name": "advanced",
+ "icon": "images/advanced.png",
+ "description": "Advanced",
+ "params": [
+ {
+ "name": "Search Queries",
+ "type": "text",
+ "value": "",
+ }
+ ],
+ }
+ controls.append(advanced_control)
+
+ return {
+ "controls": controls,
+ }
@gen.coroutine
def get(self):
diff --git a/start_app.sh b/start_app.sh
index 3f03e56..a8b6a0a 100755
--- a/start_app.sh
+++ b/start_app.sh
@@ -3,7 +3,7 @@
# This script runs the Curation application
#######################################################################################################################
# DEFAULT VARIABLES
-INGESTION="object,face"
+INGESTION="object" #,face"
EXP_TYPE=compose
DEBUG="0"
DEVICE="CPU"
diff --git a/udf/Dockerfile b/udf/Dockerfile
index 8114b5c..32d3cc1 100644
--- a/udf/Dockerfile
+++ b/udf/Dockerfile
@@ -9,8 +9,9 @@ ENV PYTHONDONTWRITEBYTECODE=1 \
PYTHONUNBUFFERED=1
# fixes CVE-2023-4911 vulnerability on Ubuntu 22.04
+# hadolint ignore=DL3008
RUN apt-get update && \
- apt-get install --only-upgrade libc-bin libc6 && \
+ apt-get install -y --only-upgrade --no-install-recommends libc-bin libc6 && \
apt-get install -y -q --no-install-recommends python3-pip python3-venv \
curl libgl1-mesa-glx && rm -rf /var/lib/apt/lists/*
RUN python3 -m venv ${VIRTUAL_ENV}
diff --git a/udf/requirements.txt b/udf/requirements.txt
index ef85742..38d54e7 100644
--- a/udf/requirements.txt
+++ b/udf/requirements.txt
@@ -198,9 +198,9 @@ tomli==1.1.0 \
vdms==0.0.23 \
--hash=sha256:2dec153f8cb33f27cdb9ab33125198195fd29a6777cf890fc735774ed9327874 \
--hash=sha256:7cd93242df644947c009559f31b4e19eb1dab6b62ab783ce5a2a54a4ad392f57
-werkzeug==3.1.6 \
- --hash=sha256:210c6bede5a420a913956b4791a7f4d6843a43b6fcee4dfa08a65e93007d0d25 \
- --hash=sha256:7ddf3357bb9564e407607f988f683d72038551200c704012bb9a4c523d42f131
+werkzeug==3.1.8 \
+ --hash=sha256:63a77fb8892bf28ebc3178683445222aa500e48ebad5ec77b0ad80f8726b1f50 \
+ --hash=sha256:9bad61a4268dac112f1c5cd4630a56ede601b6ed420300677a869083d70a4c44
wheel==0.46.3 \
--hash=sha256:4b399d56c9d9338230118d705d9737a2a468ccca63d5e813e2a4fc7815d8bc4d \
--hash=sha256:e3e79874b07d776c40bd6033f8ddf76a7dad46a7b8aa1b2787a83083519a1803
diff --git a/video/Dockerfile b/video/Dockerfile
index 4193004..ad5732d 100644
--- a/video/Dockerfile
+++ b/video/Dockerfile
@@ -1,59 +1,48 @@
-FROM openvisualcloud/xeon-ubuntu2204-media-nginx:23.1@sha256:d19eb597dc210134063803630ae2ea1ec84dfd4189138f59551e2f5ed047284a as build
+FROM openvisualcloud/xeon-ubuntu2204-media-nginx:23.1@sha256:d19eb597dc210134063803630ae2ea1ec84dfd4189138f59551e2f5ed047284a
ENV VIRTUAL_ENV=/opt/venv
ARG DEBIAN_FRONTEND=noninteractive
-# fixes CVE-2023-4911 vulnerability on Ubuntu 22.04
+# Prevent Python from writing .pyc files and enable unbuffered logging
+ENV PYTHONDONTWRITEBYTECODE=1 \
+ PYTHONUNBUFFERED=1
+
+# hadolint ignore=DL3008
RUN apt-get update && \
- apt-get install --only-upgrade libc-bin libc6 && \
+ apt-get install -y --only-upgrade --no-install-recommends libc-bin libc6 && \
apt-get install -y -q --no-install-recommends python3-setuptools \
python3-tornado python3-ply python3-pip python3-venv \
&& rm -rf /var/lib/apt/lists/*
+
RUN python3 -m venv ${VIRTUAL_ENV}
+
+# COPY --from=lcc_base_video_image:latest ${VIRTUAL_ENV} ${VIRTUAL_ENV}
+# COPY --from=lcc_base_video_image:latest /home /home
+
+# activate virtual environment
ENV PATH="$VIRTUAL_ENV/bin:$PATH"
-COPY resources /home/resources
+# Set the working directory in the container
+WORKDIR /home
+
COPY requirements.* /home/
ARG DEVICE="CPU"
ENV DEVICE="${DEVICE}"
-RUN pip3 install --no-cache-dir --require-hashes -r /home/requirements.txt --index-url https://download.pytorch.org/whl/cpu --extra-index-url https://pypi.org/simple
-RUN if [ "${DEVICE}" = "CPU" ]; then \
- pip3 install --no-cache-dir --require-hashes -r /home/requirements.${DEVICE}.txt --index-url https://download.pytorch.org/whl/cpu --extra-index-url https://pypi.org/simple; \
- else \
- pip3 install --no-cache-dir --require-hashes -r /home/requirements.${DEVICE}.txt ; \
- fi;
-
-RUN omz_downloader --list /home/resources/models/models.lst -o /home/resources/models --precisions FP16
+# RUN pip3 install --no-cache-dir -r /home/requirements.in
+RUN pip3 install --no-cache-dir --require-hashes -r /home/requirements.txt
-# FINAL IMAGE
-FROM openvisualcloud/xeon-ubuntu2204-media-nginx:23.1@sha256:d19eb597dc210134063803630ae2ea1ec84dfd4189138f59551e2f5ed047284a
-ENV VIRTUAL_ENV=/opt/venv
-ARG DEBIAN_FRONTEND=noninteractive
-
-# Prevent Python from writing .pyc files and enable unbuffered logging
-ENV PYTHONDONTWRITEBYTECODE=1 \
- PYTHONUNBUFFERED=1
-
-RUN apt-get update && \
- apt-get install --only-upgrade libc-bin libc6 && \
- apt-get install -y -q --no-install-recommends python3-tornado \
- curl libgl1-mesa-glx && rm -rf /var/lib/apt/lists/*
-ENV PATH="$VIRTUAL_ENV/bin:$PATH"
-
-# COPY --from=build /usr/local/lib/python3.10 /usr/local/lib/python3.10
-COPY --from=build ${VIRTUAL_ENV} ${VIRTUAL_ENV}
-COPY --from=build /home /home
-
-
-WORKDIR /home
COPY *.py /home/
+COPY *.yaml /home/
+COPY *.conf /etc/nginx/
COPY cleanup.sh /home/
COPY manage.sh /home/
-COPY *.conf /etc/nginx/
-CMD ["/bin/bash","-c","python3 /home/resources/models/download_yolo.py && /home/manage.sh"]
+# COPY frontend /home/frontend
+# COPY include /home/include
+# CMD ["/bin/bash","-c","python /home/resources/models/download_yolo.py && /home/manage.sh"]
+CMD ["/bin/bash","-c","/home/manage.sh"]
####
ARG USER
@@ -66,7 +55,7 @@ RUN if [ ${GID} -gt 0 ]; then groupadd -f -g ${GID} ${GROUP}; fi; \
if [ ${UID} -gt 0 ]; then useradd --no-log-init -d /home/${USER} -g ${GID} -K UID_MAX=${UID} -K UID_MIN=${UID} ${USER}; fi; \
touch /var/run/nginx.pid && \
mkdir -p /var/log/nginx /var/lib/nginx /var/www/cache /var/www/gen /var/www/mp4 && \
- chown -R ${UID}.${GID} /var/run/nginx.pid /var/log/nginx /var/lib/nginx /var/www /etc/nginx/nginx.conf /home/resources
+ chown -R ${UID}.${GID} /var/run/nginx.pid /var/log/nginx /var/lib/nginx /var/www /etc/nginx/nginx.conf /home
VOLUME ["/var/www"]
USER ${UID}
####
diff --git a/video/manage.sh b/video/manage.sh
index ac4247d..9d1ffc1 100755
--- a/video/manage.sh
+++ b/video/manage.sh
@@ -1,7 +1,8 @@
#!/bin/bash -e
# Watch directory
-python3 /home/watch_and_send2vdms.py ${WATCH_DIR} &
+echo "WATCH_DIR: ${WATCH_DIR}"
+python3 /home/source_watcher.py ${WATCH_DIR} &
# run tornado
-exec /home/manage.py
+exec ${VIRTUAL_ENV}/bin/python /home/manage.py
diff --git a/video/nginx.conf b/video/nginx.conf
index bfc5853..357d154 100644
--- a/video/nginx.conf
+++ b/video/nginx.conf
@@ -8,6 +8,19 @@ events {
}
http {
+
+ # Include MIME types
+ include mime.types;
+ default_type application/octet-stream;
+
+ # Basic logging
+ log_format main '$remote_addr - $remote_user [$time_local] "$request" '
+ '$status $body_bytes_sent "$http_referer" '
+ '"$http_user_agent" "$http_x_forwarded_for"';
+
+ access_log /var/log/nginx/access.log main;
+ error_log /var/log/nginx/error.log warn;
+
# `proxy_read_timeout` maximum time Nginx will wait for a response from a proxied server.
# If the proxied server does not send any data within this time, Nginx will close the connection.
# Default: 60
@@ -22,15 +35,16 @@ http {
# Default: 60
# proxy_connect_timeout 300s;
- include mime.types;
- default_type application/octet-stream;
+ # include mime.types;
+ # default_type application/octet-stream;
server {
listen 8080;
server_name _;
# client_body_timeout 10s;
# client_header_timeout 10s;
- client_max_body_size 1024M;
+ # client_max_body_size 1024M;
+ client_max_body_size 2G; # 8K frames are huge
sendfile on;
location /api/ {
@@ -46,5 +60,21 @@ http {
internal;
root /var/www;
}
+
+ # location / {
+ # # proxy_http_version 1.1;
+
+ # # Mandatory for WebSocket support
+ # proxy_set_header Upgrade $http_upgrade;
+ # proxy_set_header Connection "upgrade";
+
+ # # Standard proxy headers
+ # proxy_set_header Host $host;
+ # proxy_set_header X-Real-IP $remote_addr;
+ # proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+ # proxy_set_header X-Forwarded-Proto $scheme;
+ # proxy_read_timeout 86400;
+ # }
}
}
+
diff --git a/video/process_stream.py b/video/process_stream.py
deleted file mode 100644
index eca85bf..0000000
--- a/video/process_stream.py
+++ /dev/null
@@ -1,988 +0,0 @@
-import gc
-import multiprocessing as mp
-import os
-import queue
-import shlex
-import subprocess
-import sys
-import time # time library
-import traceback
-from concurrent.futures import ThreadPoolExecutor, as_completed
-from datetime import datetime
-from pathlib import Path
-from random import randint
-
-import cv2 # OpenCV library
-from openvino.runtime import Core
-from ultralytics import YOLO
-from ultralytics.utils.checks import check_imgsz
-from utils import str2bool
-
-import vdms
-
-
-def _sort_dict_by_frame(in_dict):
- def _by_int(key):
- return tuple(int(k) for k in key.split("_"))
-
- return dict(sorted(in_dict.items(), key=lambda x: _by_int(x[0])))
-
-
-""" GENERAL VARIABLES """
-CODE_DIR = os.getenv("CODE_DIR", "/home")
-CUSTOM_MODEL_FLAG = str2bool(os.getenv("CUSTOM_MODEL_FLAG", False))
-DBHOST = os.getenv("DBHOST", "vdms-service")
-DEBUG = os.getenv("DEBUG", "0")
-DEVICE = os.getenv("DEVICE", "CPU")
-INGESTION = os.getenv("INGESTION", "object,face")
-RESIZE_FLAG = str2bool(os.getenv("RESIZE_FLAG", False))
-OMIT_DETECTIONS_FLAG = str2bool(os.getenv("OMIT_DETECTIONS_FLAG", False))
-SHARED_OUTPUT = os.getenv("SHARED_OUTPUT", "/var/www/mp4")
-Path(SHARED_OUTPUT).mkdir(parents=True, exist_ok=True)
-TEST_MODE = str2bool(os.getenv("TEST_FLAG", False))
-TMP_LOCATION = os.getenv("TMP_LOCATION", "/var/www/cache/")
-UDF_HOST = os.getenv("UDF_HOST", "video-service")
-MODEL_NAME = os.getenv("MODEL_NAME", "yolo11n")
-
-LOCKTIMEOUT_RETRIES = 5
-ERR_KEYWORDS = [
- "timeout",
- "null search iterator",
- "outoftransactions",
- "internal server",
-]
-
-BATCH_SIZE = 1
-DBPORT = 55555
-DEBUG_FLAG = True if DEBUG == "1" else False
-DETECTION_THRESHOLD = 0.25
-DEVICE_OV = "AUTO"
-HALF_FLAG = True
-IOU_THRESHOLD = 0.7
-MODEL_PRECISION = "FP16"
-MODEL_W, MODEL_H = (640, 640)
-TARGET_FPS = 15
-UDF_PORT = 5011
-WRITER_FOURCC = cv2.VideoWriter_fourcc(*"mp4v") # avc1, mp4v
-NUM_USUABLE_CPUS = 2 # os.cpu_count()
-
-if CUSTOM_MODEL_FLAG:
- model_path = f"{CODE_DIR}/resources/models/ultralytics/custom_models/{MODEL_NAME}"
-else:
- model_path = f"{CODE_DIR}/resources/models/ultralytics/{MODEL_NAME}/{MODEL_PRECISION}/{MODEL_NAME}"
-
-if not TEST_MODE:
- db = vdms.vdms()
- db.connect(DBHOST, DBPORT)
-
-if DEVICE == "GPU":
- model_path += ".engine"
- batch_size = int(os.environ.get("GPU_BATCH_SIZE", 1))
- os.environ["CUDA_VISIBLE_DEVICES"] = "0"
- from torch.cuda import empty_cache
-else:
- model_path += "_openvino_model/"
- batch_size = int(os.environ.get("CPU_BATCH_SIZE", 1)) # 8
-
-device_input = DEVICE.lower() if DEVICE == "CPU" else "cuda"
-
-all_metadata = {}
-send_metadata_queue = mp.Queue()
-
-
-def retry_query(query, num_retries: int = LOCKTIMEOUT_RETRIES, sleep_timer: int = 0):
- global db
- for ridx in range(num_retries + 1):
- response, _ = db.query(query, [[]])
- if "FailedCommand" in response[0] and any(
- k in response[0]["info"].lower() for k in ERR_KEYWORDS
- ):
- err = response[0]["info"]
- if DEBUG == "1":
- query_type = list(query[0].keys())[0]
- print(
- f"DEBUG [process_stream Attempt #{ridx}] Received '{err}' for {query_type} query",
- flush=True,
- )
- if sleep_timer > 0:
- time.sleep(sleep_timer)
- else:
- if DEBUG == "1":
- print(
- f"[DEBUG process_stream] Successful query response: {response}",
- flush=True,
- )
- break # Continue
- return response
-
-
-def load_models():
- # OBJECT DETECTION
- model = YOLO(model_path, verbose=False, task="detect")
-
- # FACE, AGE, GENDER, AND EMOTIONS
- ie = Core()
- face_detection_model_xml = f"{CODE_DIR}/resources/models/intel/face-detection-adas-0001/{MODEL_PRECISION}/face-detection-adas-0001.xml"
- face_detection_model = ie.read_model(
- model=face_detection_model_xml,
- weights=face_detection_model_xml.replace(".xml", ".bin"),
- )
- # face_det_w, face_det_h = 672, 384
- _, face_det_c, face_det_h, face_det_w = face_detection_model.inputs[0].shape
- face_det_compiled_model = ie.compile_model(face_detection_model, DEVICE_OV)
-
- age_gender_classification_model_xml = f"{CODE_DIR}/resources/models/intel/age-gender-recognition-retail-0013/{MODEL_PRECISION}/age-gender-recognition-retail-0013.xml"
- age_gender_classification_model = ie.read_model(
- model=age_gender_classification_model_xml,
- weights=age_gender_classification_model_xml.replace(".xml", ".bin"),
- )
- # ag_w, ag_h = 62, 62
- _, ag_c, ag_h, ag_w = age_gender_classification_model.inputs[0].shape
- ag_compiled_model = ie.compile_model(age_gender_classification_model, DEVICE_OV)
-
- emotions_classification_model_xml = f"{CODE_DIR}/resources/models/intel/emotions-recognition-retail-0003/{MODEL_PRECISION}/emotions-recognition-retail-0003.xml"
- emotions_classification_model = ie.read_model(
- model=emotions_classification_model_xml,
- weights=emotions_classification_model_xml.replace(".xml", ".bin"),
- )
- # em_w, em_h = 64, 64
- _, em_c, em_h, em_w = emotions_classification_model.inputs[0].shape
- em_compiled_model = ie.compile_model(emotions_classification_model, DEVICE_OV)
-
- return (
- model,
- [face_det_compiled_model, ag_compiled_model, em_compiled_model],
- [face_det_c, face_det_h, face_det_w],
- [ag_c, ag_h, ag_w],
- [em_c, em_h, em_w],
- )
-
-
-model, face_models, face_det_CHW, ag_CHW, em_CHW = load_models()
-
-
-""" DETECTION FUNCTIONS """
-
-
-# Extract metadata from object model results
-def extract_metadata_from_results(
- stream_name, frameNum, results, img_size, fps=TARGET_FPS
-):
- fW, fH = img_size
- metadata = dict()
- try:
- for _, result in enumerate(results):
- # GET METADATA FOR CLIP
- boxes = result.boxes.cpu()
- oidx = 0
- for box in boxes:
- confidence = float(box.conf.item())
- if confidence > DETECTION_THRESHOLD:
- class_id = int(box.cls.item())
- class_name = str(result.names[class_id])
-
- if not OMIT_DETECTIONS_FLAG:
- timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S.%f")[:-3]
- print(
- # f"[OBJECT DETECTION] {class_name} detected in frame {frameNum} (Total detected: {current_cnt})",
- f"[{timestamp}] {stream_name} DETECTION on Frame {frameNum}: {class_name} detected",
- flush=True,
- )
- x1, y1, x2, y2 = box.xyxy.tolist()[0]
- height = min(y2, fH) - max(0, y1)
- width = min(x2, fW) - max(0, x1)
- object_res = [
- x1,
- y1,
- height,
- width,
- result.names[class_id],
- confidence,
- fH,
- fW,
- ]
-
- framenum_str = f"{frameNum:04d}_{oidx:04d}"
- if DEBUG_FLAG:
- meta_str = ",".join(
- [str(o) for o in object_res + [framenum_str]]
- )
- print(f"[{stream_name} METADATA],{meta_str}", flush=True)
-
- metadata[framenum_str] = {
- "frameId": frameNum,
- "bbId": framenum_str,
- "bbox": {
- "x": int(object_res[0]),
- "y": int(object_res[1]),
- "height": int(object_res[2]),
- "width": int(object_res[3]),
- "object": str(object_res[4]),
- "object_det": {
- "confidence": float(object_res[5]),
- "frameH": int(fH),
- "frameW": int(fW),
- },
- },
- }
- oidx += 1
-
- except Exception:
- e = traceback.format_exc()
- print(f"Error in {stream_name} extract_metadata_from_results: {e}", flush=True)
-
- return metadata
-
-
-# Detect faces from frame
-def face_detection(
- stream_name, frameNum, frame, img_size, face_models, face_det_CHW, ag_CHW, em_CHW
-):
- face_det_compiled_model, ag_compiled_model, em_compiled_model = face_models
- face_det_c, face_det_h, face_det_w = face_det_CHW
- ag_c, ag_h, ag_w = ag_CHW
- em_c, em_h, em_w = em_CHW
- W, H = img_size
- bs = 1
-
- genders = ["female", "male"]
- emotions = ["neutral", "happy", "sad", "surprise", "anger"]
-
- # Resize expects HWC
- input_image = cv2.resize(
- frame, (face_det_w, face_det_h), interpolation=cv2.INTER_AREA
- )
- input_image = input_image.transpose(2, 0, 1) # Shape: CHW
- input_image = input_image.reshape((bs, face_det_c, face_det_h, face_det_w))
-
- output_layer = face_det_compiled_model.output(0)
- result = face_det_compiled_model([input_image])[output_layer]
-
- # Process the detections
- faces = []
- metadata = dict()
- oidx = 1
- for detection in result[0][0]:
- confidence = float(detection[2])
- if confidence > DETECTION_THRESHOLD:
- # Draw a bounding box around the face
- x1 = int(detection[3] * frame.shape[1])
- if x1 < 0:
- x1 = 0
-
- y1 = int(detection[4] * frame.shape[0])
- if y1 < 0:
- y1 = 0
-
- x2 = int(detection[5] * frame.shape[1])
- if x2 > frame.shape[1] - 1:
- x2 = frame.shape[1] - 1
-
- y2 = int(detection[6] * frame.shape[0])
- if y2 > frame.shape[0] - 1:
- y2 = frame.shape[0] - 1
-
- height = y2 - y1
- width = x2 - x1
-
- face_roi = frame[y1:y2, x1:x2]
- # print(face_roi.shape)
- age = gender = emotion = None
- try:
- ag_face_blob = cv2.resize(
- face_roi, (ag_w, ag_h), interpolation=cv2.INTER_AREA
- )
- ag_face_blob = ag_face_blob.transpose((2, 0, 1))
- ag_face_blob = ag_face_blob.reshape((bs, ag_c, ag_h, ag_w))
- ag_result = ag_compiled_model([ag_face_blob])
- age = int(ag_result["fc3_a"].flatten()[0] * 100)
- gender = str(genders[ag_result["prob"].argmax()])
- except Exception as e:
- print(f"Error occurred: {e}. Skipping age-gender model", flush=True)
-
- try:
- em_face_blob = cv2.resize(
- face_roi, (em_w, em_h), interpolation=cv2.INTER_AREA
- )
- em_face_blob = em_face_blob.transpose((2, 0, 1))
- em_face_blob = em_face_blob.reshape((bs, em_c, em_h, em_w))
- em_result = em_compiled_model([em_face_blob])[
- em_compiled_model.output(0)
- ]
- emotion = str(emotions[em_result.argmax()])
- except Exception as e:
- print(f"Error occurred: {e}. Skipping emotion model", flush=True)
- face_res = [x1, y1, height, width, age, gender, emotion, confidence, H, W]
- # print(face_res)
- faces.append(face_res)
-
- tdict = {
- "x": int(face_res[0]),
- "y": int(face_res[1]),
- "height": int(face_res[2]),
- "width": int(face_res[3]),
- "object": "face",
- "object_det": {
- "age": int(face_res[4]),
- "gender": str(face_res[5]),
- "emotion": str(face_res[6]),
- "confidence": float(face_res[7]),
- "frameH": int(H),
- "frameW": int(W),
- },
- }
-
- framenum_str = f"{frameNum:04d}_{oidx:04d}"
- if DEBUG_FLAG:
- meta_str = ",".join([str(o) for o in face_res + [framenum_str]])
- print(f"[{stream_name} METADATA],{meta_str}", flush=True)
-
- metadata[framenum_str] = {
- "frameId": frameNum,
- "bbId": framenum_str,
- "bbox": tdict,
- }
- oidx += 1
-
- return metadata
-
-
-# Inference Function
-def infer_worker(
- stream_name,
- frameNum,
- frame,
- img_size,
- INGESTION,
- fps=TARGET_FPS,
-): # img_size:(W,H)
- global model, face_models, face_det_CHW, ag_CHW, em_CHW
-
- height, width = frame.shape[:2]
- if (width, height) != img_size:
- frame = cv2.resize(frame, img_size)
-
- metadata = {}
- metadata_face = {}
- try:
- if "object" in INGESTION:
- results = model.predict(
- frame,
- imgsz=(img_size[1], img_size[0]),
- batch=BATCH_SIZE,
- conf=DETECTION_THRESHOLD,
- iou=IOU_THRESHOLD,
- half=HALF_FLAG,
- device=device_input,
- verbose=False,
- stream=True,
- )
-
- metadata = extract_metadata_from_results(
- stream_name, frameNum, results, img_size, fps=fps
- )
- del results
-
- if "face" in INGESTION:
- metadata_face = face_detection(
- stream_name,
- frameNum,
- frame,
- img_size,
- face_models,
- face_det_CHW,
- ag_CHW,
- em_CHW,
- )
-
- if DEVICE == "GPU":
- empty_cache() # Frees memory no longer used
- gc.collect() # Forces garbage collector
- except Exception:
- e = traceback.format_exc()
- print(f"Error in {stream_name} infer_worker: {e}", flush=True)
- return metadata, metadata_face
-
-
-""" HELPFUL FUNCTIONS """
-
-
-# Manual FPS calculation if OpenCV reports 0
-def manual_fps_calculation(src, num_frames=10):
- vid_obj = cv2.VideoCapture(src)
-
- frame_count = 0
- start_t = time.time()
-
- while frame_count < num_frames:
- grabbed, frame = vid_obj.read()
-
- if not grabbed:
- break
-
- frame_count += 1
-
- end_t = time.time()
- vid_obj.release()
-
- elapsed_t = end_t - start_t
-
- if elapsed_t > 0:
- return frame_count / elapsed_t
- else:
- return 0
-
-
-# Generate and run UDF query
-def get_udf_query(
- filename_path,
- properties,
- ingest_mode,
- new_size,
- id="udf_metadata",
- metadata=None,
- test_mode=TEST_MODE,
-):
- query = {
- "AddVideo": {
- "from_file_path": str(filename_path), # from_server_file
- "is_local_file": True,
- "properties": properties,
- "operations": [
- {
- "type": "syncremoteOp", # "remoteOp",
- "url": f"http://{UDF_HOST}:{UDF_PORT}/video",
- "options": {
- "id": id,
- "otype": ingest_mode,
- "media_type": "video",
- "input_sizeWH": new_size,
- "filename": properties["Name"],
- "ingestion": 1,
- },
- }
- ],
- }
- }
-
- if id == "udf_metadata" and metadata is not None:
- query["AddVideo"]["operations"][0]["options"]["metadata"] = metadata
-
- if test_mode:
- return
-
- filename = str(Path(filename_path).name)
- if DEBUG_FLAG:
- print(
- f"[TIMING],start_udf_ingest_{ingest_mode},{filename}," + str(time.time()),
- flush=True,
- )
- try:
- res = retry_query([query], sleep_timer=randint(1, 5))
-
- if DEBUG_FLAG:
- print(
- f"[TIMING],end_udf_ingest_{ingest_mode},{filename}," + str(time.time()),
- flush=True,
- )
- print(f"[DEBUG] {filename} PROPERTIES: {properties}", flush=True)
- print(f"[DEBUG] {filename} INGEST_VIDEO RESPONSE: {res}", flush=True)
- except Exception:
- e = traceback.format_exc()
- print(f"[DEBUG] VDMS Query Exception: {e}", flush=True)
-
- # elapsed_time = time.time() - start_t
-
- # db.disconnect()
- # del db
-
-
-# Release Video Writer object and re-encode video to seek via ffmpeg later
-def release_clip_and_reencode(clip_key, _out_vid, clip_filename, tmp_file, target_fps):
- if DEBUG == "1":
- print(
- f"[TIMING],start_release_clip,{clip_key},{time.time()}",
- flush=True,
- )
- _out_vid.release()
- if DEBUG == "1":
- print(
- f"[TIMING],end_release_clip,{clip_key},{time.time()}",
- flush=True,
- )
- _out_vid = None
-
- # Re-encode video in order to seek via ffmpeg later
- GENERAL_OPTS = "-flags -global_header -hide_banner -loglevel error -nostats -tune zerolatency -flush_packets 0" # -filter:v fps={target_fps}
- CONVERSION = f"-c:v libx264 -preset ultrafast -filter:v fps=fps={target_fps}" # "-c:v libx264 -preset medium"
- reencode_cmd = f"ffmpeg -y -i {tmp_file} {GENERAL_OPTS} {CONVERSION} -crf 23 -c:a copy {clip_filename}"
- cmd_list = shlex.split(reencode_cmd)
- if DEBUG == "1":
- print(
- f"[TIMING],start_reencode,{clip_key},{time.time()}",
- flush=True,
- )
- subprocess.run(cmd_list, check=True)
- end_time = time.time()
- # filename = str(Path(clip_filename).name)
- if DEBUG == "1":
- print(
- f"[TIMING],end_reencode,{clip_key},{end_time}",
- flush=True,
- )
- print(f"[TIMING],Save clip,{clip_key},{end_time}", flush=True)
- os.remove(tmp_file)
- return _out_vid
-
-
-# method to send metadata to VDMS once clip is saved
-def metadata2vdms(
- clip_key,
- clip_filename,
- clip_metadata,
- width,
- height,
-):
- if DEBUG == "1":
- print(
- f"[TIMING],start_clip_metadata,{clip_key},{time.time()}",
- flush=True,
- )
-
- # Send metadata to UDF
- properties = {
- "Name": clip_key, # .split("/")[-1],
- "category": "video_path_rop",
- }
-
- combined_metadata = clip_metadata["object"] if "object" in clip_metadata else {}
- if "face" in clip_metadata:
- for face_frameidx_bbidx, value in clip_metadata["face"].items():
- face_frameidx, face_bbidx = face_frameidx_bbidx.split("_")
- max_obj_idx = 0
- for obj_frameidx_bbidx in combined_metadata:
- if face_frameidx in obj_frameidx_bbidx:
- _, obj_bbidx_ = obj_frameidx_bbidx.split("_")
- max_obj_idx = max(max_obj_idx, int(obj_bbidx_))
-
- if max_obj_idx > 0:
- new_face_bbidx = max_obj_idx + 1
- new_key = f"{face_frameidx}_{new_face_bbidx:04d}"
- combined_metadata[new_key] = value
- combined_metadata[new_key]["bbId"] = new_key
- else:
- combined_metadata[face_frameidx_bbidx] = value
-
- combined_metadata = _sort_dict_by_frame(combined_metadata)
- get_udf_query(
- clip_filename,
- properties,
- INGESTION.replace(",", "+"),
- (width, height),
- id="udf_metadata",
- metadata=combined_metadata,
- test_mode=TEST_MODE,
- )
-
- if DEBUG == "1":
- print(
- f"[TIMING],end_clip_metadata,{clip_key},{time.time()}",
- flush=True,
- )
-
-
-""" CLASSES """
-
-
-# method to save clip
-def save_clip(
- clip_filename, clip_id, tmp_file, _out_vid, frame_count, frameNum, target_fps
-):
- clip_key = Path(clip_filename).name
- if DEBUG == "1":
- print(
- f"[DEBUG] Clip {clip_key} (clip_id: {clip_id}) contains {frame_count} frames",
- flush=True,
- )
- _out_vid = release_clip_and_reencode(
- clip_key,
- _out_vid,
- clip_filename,
- tmp_file,
- target_fps,
- )
- if DEBUG == "1":
- print(
- f"[TIMING],end_get_clips,{clip_key},{time.time()}",
- flush=True,
- )
- return _out_vid
-
-
-# method to create clips (read frame write to file; add name to list)
-def send_metadata():
- global all_metadata
- clip_filename = ""
- clip_key = ""
- width = 0
- height = 0
- while True:
- try:
- queue_details = send_metadata_queue.get()
- if queue_details is None:
- break
-
- (clip_key, clip_filename, width, height) = queue_details
-
- metadata2vdms(
- clip_key,
- clip_filename,
- all_metadata[clip_key],
- width,
- height,
- )
- del all_metadata[clip_key]
-
- except queue.Empty:
- pass
-
-
-# defining a helper class for implementing multi-threading
-class VideoStream:
- # initialization method
- def __init__(self, src, fps=TARGET_FPS, fourcc=WRITER_FOURCC, camera_name=None):
- self.stream_id = src
- self.fourcc = fourcc
-
- if "://" in str(self.stream_id):
- if camera_name is not None:
- self.stream_name = camera_name
- else:
- self.stream_name = str(self.stream_id).split("/")[-1]
- else:
- self.stream_name = Path(self.stream_id).stem
-
- if self.stream_id.startswith("rtsp"):
- os.environ["OPENCV_FFMPEG_CAPTURE_OPTIONS"] = "rtsp_transport;tcp"
-
- # Check that object is opened successfully
- self.connect_to_stream(time_limit_mins=5)
- if DEBUG == "1":
- print(
- f"[TIMING],Start processing,{self.stream_name},{time.time()}",
- flush=True,
- )
-
- self.setup_stream(fps)
-
- # Create ThreadPoolExecutor
- self.executor = ThreadPoolExecutor(max_workers=NUM_USUABLE_CPUS)
-
- # method to start thread
- def start(self):
- self.stopped = False
- self.t = []
- self.t.append(
- self.executor.submit(
- self.get_frames,
- )
- )
- self.t.append(
- self.executor.submit(
- send_metadata,
- )
- )
-
- # method to stop reading frames
- def stop(self):
- for t in as_completed(self.t):
- try:
- _ = t.result()
- except Exception as t_e:
- print(f"[DEBUG] Exception occurred in thread: {t_e}")
-
- # self.stopped = True
- self.video_obj.release()
-
- # method to open stream/video within 5min (default) limit
- def connect_to_stream(self, time_limit_mins=5):
- # opening video capture stream
- self.video_obj = cv2.VideoCapture(self.stream_id, cv2.CAP_FFMPEG)
-
- stream_available = False
- time_limit_secs = time_limit_mins * 60
- connect_time = time.time()
- while not stream_available:
- if self.video_obj.isOpened():
- stream_available = True
- elif self.stream_id.startswith("rtsp"):
- if time.time() - connect_time < time_limit_secs:
- self.video_obj = cv2.VideoCapture(self.stream_id, cv2.CAP_FFMPEG)
- else:
- print(
- f"Exceeds {time_limit_mins} mins limit to connect to {self.stream_name}. Exiting ..."
- )
- exit(1)
-
- # Gets video fps and framecount
- def get_fps_and_framecnt(self, fps):
- self.input_fps = int(self.video_obj.get(cv2.CAP_PROP_FPS)) # hardware fps
- if self.input_fps == 0: # Case when FPs isn't available
- self.input_fps = manual_fps_calculation(self.stream_id, num_frames=10)
-
- self.target_fps = fps if self.input_fps > fps else self.input_fps
- self.frame_skip = int(self.input_fps / self.target_fps)
- if self.frame_skip < 1:
- self.frame_skip = 1
-
- print(f"FPS of {self.stream_name} input stream: {self.input_fps}", flush=True)
- print(f"FPS of {self.stream_name} output mp4: {self.target_fps}", flush=True)
-
- # Frame count for videos
- self.frame_count = None
- if "://" not in str(self.stream_id):
- self.frame_count = int(self.video_obj.get(cv2.CAP_PROP_FRAME_COUNT))
-
- # Gets frame W and H details
- def get_frameWH(self):
- input_width = int(self.video_obj.get(cv2.CAP_PROP_FRAME_WIDTH))
- input_height = int(self.video_obj.get(cv2.CAP_PROP_FRAME_HEIGHT))
-
- if RESIZE_FLAG or ((input_height * input_width) < (MODEL_H * MODEL_W)):
- new_sizeHW = check_imgsz([MODEL_H, MODEL_W]) # expects hxw
- else:
- new_sizeHW = check_imgsz([input_height, input_width]) # expects hxw
-
- new_sizeWH = (new_sizeHW[1], new_sizeHW[0])
-
- self.width = new_sizeWH[0]
- self.height = new_sizeWH[1]
-
- # Sets up important info for stream
- def setup_stream(self, fps):
- self.inference_queue = mp.Queue()
- self.retrieved_frames = 0
- self.num_frames_processed = 0
-
- self.get_fps_and_framecnt(fps)
-
- self.get_frameWH()
-
- self._out_vid = None
- self.clip_end_frame = {}
- self.clip_filename = ""
- self.clip_frame_count = 0
- self.clip_frame_inds = []
- self.clip_id = 0
- self.clip_length_in_secs = 10
- self.clip_total_frames = int(float(self.clip_length_in_secs * self.target_fps))
-
- # method to process a frame
- def get_frames(self):
- clip_frame_idx = 0
- clip_id = 0
- if DEBUG == "1":
- print(
- f"[TIMING],start_get_frames,{self.stream_name},{time.time()}",
- flush=True,
- )
- while True:
- grabbed, frame = self.video_obj.read() # Read next frame
-
- if not grabbed: # or self.stopped:
- # self.stopped = True
- self.inference_queue.put(None)
- break
-
- frameNum = int(self.video_obj.get(cv2.CAP_PROP_POS_FRAMES))
- skip_frame_num = (frameNum - 1) % self.frame_skip
-
- if clip_frame_idx % self.clip_total_frames == 0:
- if "://" not in str(self.stream_id):
- clip_filename = f"{SHARED_OUTPUT}/{self.stream_name}_{clip_id}.mp4"
- else:
- clip_filename = (
- f"{SHARED_OUTPUT}/{self.stream_name}_{time.time()}.mp4"
- )
-
- tmp_file = TMP_LOCATION + clip_filename.split("/")[-1]
-
- if skip_frame_num == 0:
- h, w = frame.shape[:2]
- if (w, h) != (self.width, self.height):
- frame = cv2.resize(frame, (self.width, self.height))
-
- queue_details = (
- frameNum, # Overall frame number
- (clip_frame_idx % self.clip_total_frames)
- + 1, # Frame index in clip [1-indexed like opencv]
- clip_id, # Clip number
- clip_filename,
- tmp_file,
- frame.copy(), # Frame
- )
-
- self.inference_queue.put(queue_details)
- self.retrieved_frames += 1
-
- clip_frame_idx += 1
- if clip_frame_idx % self.clip_total_frames == 0:
- clip_id += 1
-
- if DEBUG == "1":
- print(
- f"[TIMING],end_get_frames,{self.stream_name},{time.time()}", flush=True
- )
-
-
-def process_stream(camera_src, camera_name=None):
- global all_metadata
- webcam_stream = VideoStream(str(camera_src), camera_name=camera_name)
- if DEBUG == "1":
- print(
- f"[TIMING],Start processing,{webcam_stream.stream_name},{time.time()}",
- flush=True,
- )
-
- start = time.time()
- # Start retrieving frames and add to queue
- webcam_stream.start()
- _out_vid = None
- clip_filename = ""
- clip_frame_idx = 0
- clip_id = 0
- clip_key = ""
- frameNum = 0
- tmp_file = ""
- while True:
- queue_details = webcam_stream.inference_queue.get()
-
- if queue_details is None:
- if _out_vid is not None:
- frame_count = clip_frame_idx
- if frame_count > TARGET_FPS:
- _out_vid = save_clip(
- clip_filename,
- clip_id,
- tmp_file,
- _out_vid,
- frame_count,
- frameNum,
- webcam_stream.target_fps,
- )
- send_metadata_queue.put(
- (
- clip_key,
- clip_filename,
- webcam_stream.width,
- webcam_stream.height,
- )
- )
- else:
- _out_vid = None
- send_metadata_queue.put(None)
- print("End of stream")
- break
-
- frameNum, clip_frame_idx, clip_id, clip_filename, tmp_file, frame = (
- queue_details
- )
- clip_key = Path(clip_filename).name
- if DEBUG == "1":
- print(
- f"[TIMING],start_infer_worker,{clip_key}-{clip_frame_idx},{time.time()}",
- flush=True,
- )
-
- metadata, metadata_face = infer_worker(
- webcam_stream.stream_name,
- clip_frame_idx,
- frame,
- (webcam_stream.width, webcam_stream.height), # img_size,
- INGESTION,
- fps=webcam_stream.target_fps,
- )
-
- if DEBUG == "1":
- print(
- f"[TIMING],end_infer_worker,{clip_key}-{clip_frame_idx},{time.time()}",
- flush=True,
- )
-
- all_metadata.setdefault(clip_key, {})
- all_metadata[clip_key].setdefault("object", {})
- all_metadata[clip_key]["object"].update(metadata)
- all_metadata[clip_key].setdefault("face", {})
- all_metadata[clip_key]["face"].update(metadata_face)
- webcam_stream.num_frames_processed += 1
-
- if clip_frame_idx - 1 == 0:
- if DEBUG == "1":
- print(
- f"[TIMING],start_get_clips,{clip_key},{time.time()}",
- flush=True,
- )
- _out_vid = cv2.VideoWriter(
- tmp_file,
- fourcc=webcam_stream.fourcc,
- fps=webcam_stream.target_fps,
- frameSize=(webcam_stream.width, webcam_stream.height),
- )
- if DEBUG == "1":
- print(
- f"[TIMING],Start new clip,{clip_key},{time.time()}",
- flush=True,
- )
-
- _out_vid.write(frame)
-
- if clip_frame_idx == webcam_stream.clip_total_frames:
- frame_count = clip_frame_idx
- _out_vid = save_clip(
- clip_filename,
- clip_id,
- tmp_file,
- _out_vid,
- frame_count,
- frameNum,
- webcam_stream.target_fps,
- )
- queue_details = (
- clip_key,
- clip_filename,
- webcam_stream.width,
- webcam_stream.height,
- )
- send_metadata_queue.put(queue_details)
-
- webcam_stream.stop()
-
- end = time.time()
-
- # printing time elapsed and fps
- if DEBUG == "1":
- elapsed = end - start
- print(
- "[DEBUG] Stream name:{}, FPS: {} , Elapsed Time: {}, Num. Retrieved Frames: {}, Num. Processed Frames: {}".format(
- webcam_stream.stream_name,
- webcam_stream.target_fps,
- elapsed,
- webcam_stream.retrieved_frames,
- webcam_stream.num_frames_processed,
- ),
- flush=True,
- )
-
- print(
- f"[TIMING],Completed processing,{webcam_stream.stream_name},{end}",
- flush=True,
- )
-
-
-""" MAIN FUNCTION """
-
-if __name__ == "__main__":
- camera_src = sys.argv[1]
- camera_name = sys.argv[2] if len(sys.argv) == 3 else None
-
- process_stream(str(camera_src), camera_name)
diff --git a/video/requirements.CPU.txt b/video/requirements.CPU.txt
deleted file mode 100644
index 4828f07..0000000
--- a/video/requirements.CPU.txt
+++ /dev/null
@@ -1,1161 +0,0 @@
-about-time==4.2.1 \
- --hash=sha256:6a538862d33ce67d997429d14998310e1dbfda6cb7d9bbfbf799c4709847fece \
- --hash=sha256:8bbf4c75fe13cbd3d72f49a03b02c5c7dca32169b6d49117c257e7eb3eaee341
-alive-progress==3.3.0 \
- --hash=sha256:457dd2428b48dacd49854022a46448d236a48f1b7277874071c39395307e830c \
- --hash=sha256:63dd33bb94cde15ad9e5b666dbba8fedf71b72a4935d6fb9a92931e69402c9ff
-attrs==25.4.0 \
- --hash=sha256:16d5969b87f0859ef33a48b35d55ac1be6e42ae49d5e853b597db70c35c57e11 \
- --hash=sha256:adcf7e2a1fb3b36ac48d97835bb6d8ade15b8dcce26aba8bf1d14847b57a3373
-autograd==1.8.0 \
- --hash=sha256:107374ded5b09fc8643ac925348c0369e7b0e73bbed9565ffd61b8fd04425683 \
- --hash=sha256:4ab9084294f814cf56c280adbe19612546a35574d67c574b04933c7d2ecb7d78
-cffi==2.0.0 \
- --hash=sha256:00bdf7acc5f795150faa6957054fbbca2439db2f775ce831222b66f192f03beb \
- --hash=sha256:07b271772c100085dd28b74fa0cd81c8fb1a3ba18b21e03d7c27f3436a10606b \
- --hash=sha256:087067fa8953339c723661eda6b54bc98c5625757ea62e95eb4898ad5e776e9f \
- --hash=sha256:0a1527a803f0a659de1af2e1fd700213caba79377e27e4693648c2923da066f9 \
- --hash=sha256:0cf2d91ecc3fcc0625c2c530fe004f82c110405f101548512cce44322fa8ac44 \
- --hash=sha256:0f6084a0ea23d05d20c3edcda20c3d006f9b6f3fefeac38f59262e10cef47ee2 \
- --hash=sha256:12873ca6cb9b0f0d3a0da705d6086fe911591737a59f28b7936bdfed27c0d47c \
- --hash=sha256:19f705ada2530c1167abacb171925dd886168931e0a7b78f5bffcae5c6b5be75 \
- --hash=sha256:1cd13c99ce269b3ed80b417dcd591415d3372bcac067009b6e0f59c7d4015e65 \
- --hash=sha256:1e3a615586f05fc4065a8b22b8152f0c1b00cdbc60596d187c2a74f9e3036e4e \
- --hash=sha256:1f72fb8906754ac8a2cc3f9f5aaa298070652a0ffae577e0ea9bd480dc3c931a \
- --hash=sha256:1fc9ea04857caf665289b7a75923f2c6ed559b8298a1b8c49e59f7dd95c8481e \
- --hash=sha256:203a48d1fb583fc7d78a4c6655692963b860a417c0528492a6bc21f1aaefab25 \
- --hash=sha256:2081580ebb843f759b9f617314a24ed5738c51d2aee65d31e02f6f7a2b97707a \
- --hash=sha256:21d1152871b019407d8ac3985f6775c079416c282e431a4da6afe7aefd2bccbe \
- --hash=sha256:24b6f81f1983e6df8db3adc38562c83f7d4a0c36162885ec7f7b77c7dcbec97b \
- --hash=sha256:256f80b80ca3853f90c21b23ee78cd008713787b1b1e93eae9f3d6a7134abd91 \
- --hash=sha256:28a3a209b96630bca57cce802da70c266eb08c6e97e5afd61a75611ee6c64592 \
- --hash=sha256:2c8f814d84194c9ea681642fd164267891702542f028a15fc97d4674b6206187 \
- --hash=sha256:2de9a304e27f7596cd03d16f1b7c72219bd944e99cc52b84d0145aefb07cbd3c \
- --hash=sha256:38100abb9d1b1435bc4cc340bb4489635dc2f0da7456590877030c9b3d40b0c1 \
- --hash=sha256:3925dd22fa2b7699ed2617149842d2e6adde22b262fcbfada50e3d195e4b3a94 \
- --hash=sha256:3e17ed538242334bf70832644a32a7aae3d83b57567f9fd60a26257e992b79ba \
- --hash=sha256:3e837e369566884707ddaf85fc1744b47575005c0a229de3327f8f9a20f4efeb \
- --hash=sha256:3f4d46d8b35698056ec29bca21546e1551a205058ae1a181d871e278b0b28165 \
- --hash=sha256:44d1b5909021139fe36001ae048dbdde8214afa20200eda0f64c068cac5d5529 \
- --hash=sha256:45d5e886156860dc35862657e1494b9bae8dfa63bf56796f2fb56e1679fc0bca \
- --hash=sha256:4647afc2f90d1ddd33441e5b0e85b16b12ddec4fca55f0d9671fef036ecca27c \
- --hash=sha256:4671d9dd5ec934cb9a73e7ee9676f9362aba54f7f34910956b84d727b0d73fb6 \
- --hash=sha256:53f77cbe57044e88bbd5ed26ac1d0514d2acf0591dd6bb02a3ae37f76811b80c \
- --hash=sha256:5eda85d6d1879e692d546a078b44251cdd08dd1cfb98dfb77b670c97cee49ea0 \
- --hash=sha256:5fed36fccc0612a53f1d4d9a816b50a36702c28a2aa880cb8a122b3466638743 \
- --hash=sha256:61d028e90346df14fedc3d1e5441df818d095f3b87d286825dfcbd6459b7ef63 \
- --hash=sha256:66f011380d0e49ed280c789fbd08ff0d40968ee7b665575489afa95c98196ab5 \
- --hash=sha256:6824f87845e3396029f3820c206e459ccc91760e8fa24422f8b0c3d1731cbec5 \
- --hash=sha256:6c6c373cfc5c83a975506110d17457138c8c63016b563cc9ed6e056a82f13ce4 \
- --hash=sha256:6d02d6655b0e54f54c4ef0b94eb6be0607b70853c45ce98bd278dc7de718be5d \
- --hash=sha256:6d50360be4546678fc1b79ffe7a66265e28667840010348dd69a314145807a1b \
- --hash=sha256:730cacb21e1bdff3ce90babf007d0a0917cc3e6492f336c2f0134101e0944f93 \
- --hash=sha256:737fe7d37e1a1bffe70bd5754ea763a62a066dc5913ca57e957824b72a85e205 \
- --hash=sha256:74a03b9698e198d47562765773b4a8309919089150a0bb17d829ad7b44b60d27 \
- --hash=sha256:7553fb2090d71822f02c629afe6042c299edf91ba1bf94951165613553984512 \
- --hash=sha256:7a66c7204d8869299919db4d5069a82f1561581af12b11b3c9f48c584eb8743d \
- --hash=sha256:7cc09976e8b56f8cebd752f7113ad07752461f48a58cbba644139015ac24954c \
- --hash=sha256:81afed14892743bbe14dacb9e36d9e0e504cd204e0b165062c488942b9718037 \
- --hash=sha256:8941aaadaf67246224cee8c3803777eed332a19d909b47e29c9842ef1e79ac26 \
- --hash=sha256:89472c9762729b5ae1ad974b777416bfda4ac5642423fa93bd57a09204712322 \
- --hash=sha256:8ea985900c5c95ce9db1745f7933eeef5d314f0565b27625d9a10ec9881e1bfb \
- --hash=sha256:8eca2a813c1cb7ad4fb74d368c2ffbbb4789d377ee5bb8df98373c2cc0dee76c \
- --hash=sha256:92b68146a71df78564e4ef48af17551a5ddd142e5190cdf2c5624d0c3ff5b2e8 \
- --hash=sha256:9332088d75dc3241c702d852d4671613136d90fa6881da7d770a483fd05248b4 \
- --hash=sha256:94698a9c5f91f9d138526b48fe26a199609544591f859c870d477351dc7b2414 \
- --hash=sha256:9a67fc9e8eb39039280526379fb3a70023d77caec1852002b4da7e8b270c4dd9 \
- --hash=sha256:9de40a7b0323d889cf8d23d1ef214f565ab154443c42737dfe52ff82cf857664 \
- --hash=sha256:a05d0c237b3349096d3981b727493e22147f934b20f6f125a3eba8f994bec4a9 \
- --hash=sha256:afb8db5439b81cf9c9d0c80404b60c3cc9c3add93e114dcae767f1477cb53775 \
- --hash=sha256:b18a3ed7d5b3bd8d9ef7a8cb226502c6bf8308df1525e1cc676c3680e7176739 \
- --hash=sha256:b1e74d11748e7e98e2f426ab176d4ed720a64412b6a15054378afdb71e0f37dc \
- --hash=sha256:b21e08af67b8a103c71a250401c78d5e0893beff75e28c53c98f4de42f774062 \
- --hash=sha256:b4c854ef3adc177950a8dfc81a86f5115d2abd545751a304c5bcf2c2c7283cfe \
- --hash=sha256:b882b3df248017dba09d6b16defe9b5c407fe32fc7c65a9c69798e6175601be9 \
- --hash=sha256:baf5215e0ab74c16e2dd324e8ec067ef59e41125d3eade2b863d294fd5035c92 \
- --hash=sha256:c649e3a33450ec82378822b3dad03cc228b8f5963c0c12fc3b1e0ab940f768a5 \
- --hash=sha256:c654de545946e0db659b3400168c9ad31b5d29593291482c43e3564effbcee13 \
- --hash=sha256:c6638687455baf640e37344fe26d37c404db8b80d037c3d29f58fe8d1c3b194d \
- --hash=sha256:c8d3b5532fc71b7a77c09192b4a5a200ea992702734a2e9279a37f2478236f26 \
- --hash=sha256:cb527a79772e5ef98fb1d700678fe031e353e765d1ca2d409c92263c6d43e09f \
- --hash=sha256:cf364028c016c03078a23b503f02058f1814320a56ad535686f90565636a9495 \
- --hash=sha256:d48a880098c96020b02d5a1f7d9251308510ce8858940e6fa99ece33f610838b \
- --hash=sha256:d68b6cef7827e8641e8ef16f4494edda8b36104d79773a334beaa1e3521430f6 \
- --hash=sha256:d9b29c1f0ae438d5ee9acb31cadee00a58c46cc9c0b2f9038c6b0b3470877a8c \
- --hash=sha256:d9b97165e8aed9272a6bb17c01e3cc5871a594a446ebedc996e2397a1c1ea8ef \
- --hash=sha256:da68248800ad6320861f129cd9c1bf96ca849a2771a59e0344e88681905916f5 \
- --hash=sha256:da902562c3e9c550df360bfa53c035b2f241fed6d9aef119048073680ace4a18 \
- --hash=sha256:dbd5c7a25a7cb98f5ca55d258b103a2054f859a46ae11aaf23134f9cc0d356ad \
- --hash=sha256:dd4f05f54a52fb558f1ba9f528228066954fee3ebe629fc1660d874d040ae5a3 \
- --hash=sha256:de8dad4425a6ca6e4e5e297b27b5c824ecc7581910bf9aee86cb6835e6812aa7 \
- --hash=sha256:e11e82b744887154b182fd3e7e8512418446501191994dbf9c9fc1f32cc8efd5 \
- --hash=sha256:e6e73b9e02893c764e7e8d5bb5ce277f1a009cd5243f8228f75f842bf937c534 \
- --hash=sha256:f73b96c41e3b2adedc34a7356e64c8eb96e03a3782b535e043a986276ce12a49 \
- --hash=sha256:f93fd8e5c8c0a4aa1f424d6173f14a892044054871c771f8566e4008eaa359d2 \
- --hash=sha256:fc33c5141b55ed366cfaad382df24fe7dcbc686de5be719b207bb248e3053dc5 \
- --hash=sha256:fc7de24befaeae77ba923797c7c87834c73648a05a4bde34b3b7e5588973a453 \
- --hash=sha256:fe562eb1a64e67dd297ccc4f5addea2501664954f2692b69a76449ec7913ecbf
-cma==4.4.1 \
- --hash=sha256:61177b54f12bfeeac307970f8caefd8210dfa1d43a2da34e9ef8b1416d930fd8 \
- --hash=sha256:bf0621d4f52cf3354be3d0a5cd439ffed52f24f429ab02c23cf0f8ca16d427d8
-contourpy==1.3.2 \
- --hash=sha256:0475b1f6604896bc7c53bb070e355e9321e1bc0d381735421a2d2068ec56531f \
- --hash=sha256:106fab697af11456fcba3e352ad50effe493a90f893fca6c2ca5c033820cea92 \
- --hash=sha256:107ba8a6a7eec58bb475329e6d3b95deba9440667c4d62b9b6063942b61d7f16 \
- --hash=sha256:15ce6ab60957ca74cff444fe66d9045c1fd3e92c8936894ebd1f3eef2fff075f \
- --hash=sha256:1c48188778d4d2f3d48e4643fb15d8608b1d01e4b4d6b0548d9b336c28fc9b6f \
- --hash=sha256:3859783aefa2b8355697f16642695a5b9792e7a46ab86da1118a4a23a51a33d7 \
- --hash=sha256:3d80b2c0300583228ac98d0a927a1ba6a2ba6b8a742463c564f1d419ee5b211e \
- --hash=sha256:3f9e896f447c5c8618f1edb2bafa9a4030f22a575ec418ad70611450720b5b08 \
- --hash=sha256:434f0adf84911c924519d2b08fc10491dd282b20bdd3fa8f60fd816ea0b48841 \
- --hash=sha256:49b65a95d642d4efa8f64ba12558fcb83407e58a2dfba9d796d77b63ccfcaff5 \
- --hash=sha256:4caf2bcd2969402bf77edc4cb6034c7dd7c0803213b3523f111eb7460a51b8d2 \
- --hash=sha256:532fd26e715560721bb0d5fc7610fce279b3699b018600ab999d1be895b09415 \
- --hash=sha256:5ebac872ba09cb8f2131c46b8739a7ff71de28a24c869bcad554477eb089a878 \
- --hash=sha256:5f5964cdad279256c084b69c3f412b7801e15356b16efa9d78aa974041903da0 \
- --hash=sha256:65a887a6e8c4cd0897507d814b14c54a8c2e2aa4ac9f7686292f9769fcf9a6ab \
- --hash=sha256:6a37a2fb93d4df3fc4c0e363ea4d16f83195fc09c891bc8ce072b9d084853445 \
- --hash=sha256:70771a461aaeb335df14deb6c97439973d253ae70660ca085eec25241137ef43 \
- --hash=sha256:71e2bd4a1c4188f5c2b8d274da78faab884b59df20df63c34f74aa1813c4427c \
- --hash=sha256:745b57db7758f3ffc05a10254edd3182a2a83402a89c00957a8e8a22f5582823 \
- --hash=sha256:78e9253c3de756b3f6a5174d024c4835acd59eb3f8e2ca13e775dbffe1558f69 \
- --hash=sha256:82199cb78276249796419fe36b7386bd8d2cc3f28b3bc19fe2454fe2e26c4c15 \
- --hash=sha256:8b7fc0cd78ba2f4695fd0a6ad81a19e7e3ab825c31b577f384aa9d7817dc3bef \
- --hash=sha256:8c5acb8dddb0752bf252e01a3035b21443158910ac16a3b0d20e7fed7d534ce5 \
- --hash=sha256:8c942a01d9163e2e5cfb05cb66110121b8d07ad438a17f9e766317bcb62abf73 \
- --hash=sha256:8d2e74acbcba3bfdb6d9d8384cdc4f9260cae86ed9beee8bd5f54fee49a430b9 \
- --hash=sha256:90df94c89a91b7362e1142cbee7568f86514412ab8a2c0d0fca72d7e91b62912 \
- --hash=sha256:970e9173dbd7eba9b4e01aab19215a48ee5dd3f43cef736eebde064a171f89a5 \
- --hash=sha256:977e98a0e0480d3fe292246417239d2d45435904afd6d7332d8455981c408b85 \
- --hash=sha256:9be002b31c558d1ddf1b9b415b162c603405414bacd6932d031c5b5a8b757f0d \
- --hash=sha256:ad687a04bc802cbe8b9c399c07162a3c35e227e2daccf1668eb1f278cb698631 \
- --hash=sha256:b4f54d6a2defe9f257327b0f243612dd051cc43825587520b1bf74a31e2f6ef2 \
- --hash=sha256:b6945942715a034c671b7fc54f9588126b0b8bf23db2696e3ca8328f3ff0ab54 \
- --hash=sha256:b7cd50c38f500bbcc9b6a46643a40e0913673f869315d8e70de0438817cb7773 \
- --hash=sha256:ba38e3f9f330af820c4b27ceb4b9c7feee5fe0493ea53a8720f4792667465934 \
- --hash=sha256:c440093bbc8fc21c637c03bafcbef95ccd963bc6e0514ad887932c18ca2a759a \
- --hash=sha256:c49f73e61f1f774650a55d221803b101d966ca0c5a2d6d5e4320ec3997489441 \
- --hash=sha256:c66c4906cdbc50e9cba65978823e6e00b45682eb09adbb78c9775b74eb222422 \
- --hash=sha256:c6c4639a9c22230276b7bffb6a850dfc8258a2521305e1faefe804d006b2e532 \
- --hash=sha256:c85bb486e9be652314bb5b9e2e3b0d1b2e643d5eec4992c0fbe8ac71775da739 \
- --hash=sha256:cc829960f34ba36aad4302e78eabf3ef16a3a100863f0d4eeddf30e8a485a03b \
- --hash=sha256:cdd22595308f53ef2f891040ab2b93d79192513ffccbd7fe19be7aa773a5e09f \
- --hash=sha256:d0e589ae0d55204991450bb5c23f571c64fe43adaa53f93fc902a84c96f52fe1 \
- --hash=sha256:d14f12932a8d620e307f715857107b1d1845cc44fdb5da2bc8e850f5ceba9f87 \
- --hash=sha256:d32530b534e986374fc19eaa77fcb87e8a99e5431499949b828312bdcd20ac52 \
- --hash=sha256:d6658ccc7251a4433eebd89ed2672c2ed96fba367fd25ca9512aa92a4b46c4f1 \
- --hash=sha256:d91a3ccc7fea94ca0acab82ceb77f396d50a1f67412efe4c526f5d20264e6ecd \
- --hash=sha256:dc41ba0714aa2968d1f8674ec97504a8f7e334f48eeacebcaa6256213acb0989 \
- --hash=sha256:de39db2604ae755316cb5967728f4bea92685884b1e767b7c24e983ef5f771cb \
- --hash=sha256:de425af81b6cea33101ae95ece1f696af39446db9682a0b56daaa48cfc29f38f \
- --hash=sha256:ded1706ed0c1049224531b81128efbd5084598f18d8a2d9efae833edbd2b40ad \
- --hash=sha256:e1578f7eafce927b168752ed7e22646dad6cd9bca673c60bff55889fa236ebf9 \
- --hash=sha256:e259bced5549ac64410162adc973c5e2fb77f04df4a439d00b478e57a0e65512 \
- --hash=sha256:e298e7e70cf4eb179cc1077be1c725b5fd131ebc81181bf0c03525c8abc297fd \
- --hash=sha256:eab0f6db315fa4d70f1d8ab514e527f0366ec021ff853d7ed6a2d33605cf4b83 \
- --hash=sha256:f26b383144cf2d2c29f01a1e8170f50dacf0eac02d64139dcd709a8ac4eb3cfe \
- --hash=sha256:f939a054192ddc596e031e50bb13b657ce318cf13d264f095ce9db7dc6ae81c0 \
- --hash=sha256:fd93cc7f3139b6dd7aab2f26a90dde0aa9fc264dbf70f6740d498a70b860b82c
-cycler==0.12.1 \
- --hash=sha256:85cef7cff222d8644161529808465972e51340599459b8ac3ccbac5a854e0d30 \
- --hash=sha256:88bb128f02ba341da8ef447245a9e138fae777f6a23943da4540077d3601eb1c
-deprecated==1.3.1 \
- --hash=sha256:597bfef186b6f60181535a29fbe44865ce137a5079f295b479886c82729d5f3f \
- --hash=sha256:b1b50e0ff0c1fddaa5708a2c6b0a6588bb09b892825ab2b214ac9ea9d92a5223
-filelock==3.25.2 \
- --hash=sha256:b64ece2b38f4ca29dd3e810287aa8c48182bbecd1ae6e9ae126c9b35f1382694 \
- --hash=sha256:ca8afb0da15f229774c9ad1b455ed96e85a81373065fb10446672f64444ddf70
-fonttools==4.61.1 \
- --hash=sha256:0de30bfe7745c0d1ffa2b0b7048fb7123ad0d71107e10ee090fa0b16b9452e87 \
- --hash=sha256:10d88e55330e092940584774ee5e8a6971b01fc2f4d3466a1d6c158230880796 \
- --hash=sha256:11f35ad7805edba3aac1a3710d104592df59f4b957e30108ae0ba6c10b11dd75 \
- --hash=sha256:15acc09befd16a0fb8a8f62bc147e1a82817542d72184acca9ce6e0aeda9fa6d \
- --hash=sha256:17d2bf5d541add43822bcf0c43d7d847b160c9bb01d15d5007d84e2217aaa371 \
- --hash=sha256:2180f14c141d2f0f3da43f3a81bc8aa4684860f6b0e6f9e165a4831f24e6a23b \
- --hash=sha256:21e7c8d76f62ab13c9472ccf74515ca5b9a761d1bde3265152a6dc58700d895b \
- --hash=sha256:41a7170d042e8c0024703ed13b71893519a1a6d6e18e933e3ec7507a2c26a4b2 \
- --hash=sha256:41ed4b5ec103bd306bb68f81dc166e77409e5209443e5773cb4ed837bcc9b0d3 \
- --hash=sha256:497c31ce314219888c0e2fce5ad9178ca83fe5230b01a5006726cdf3ac9f24d9 \
- --hash=sha256:4c1b526c8d3f615a7b1867f38a9410849c8f4aef078535742198e942fba0e9bd \
- --hash=sha256:4d7092bb38c53bbc78e9255a59158b150bcdc115a1e3b3ce0b5f267dc35dd63c \
- --hash=sha256:4f5686e1fe5fce75d82d93c47a438a25bf0d1319d2843a926f741140b2b16e0c \
- --hash=sha256:58b0ee0ab5b1fc9921eccfe11d1435added19d6494dde14e323f25ad2bc30c56 \
- --hash=sha256:5ce02f38a754f207f2f06557523cd39a06438ba3aafc0639c477ac409fc64e37 \
- --hash=sha256:5fade934607a523614726119164ff621e8c30e8fa1ffffbbd358662056ba69f0 \
- --hash=sha256:5fe9fd43882620017add5eabb781ebfbc6998ee49b35bd7f8f79af1f9f99a958 \
- --hash=sha256:64102ca87e84261419c3747a0d20f396eb024bdbeb04c2bfb37e2891f5fadcb5 \
- --hash=sha256:664c5a68ec406f6b1547946683008576ef8b38275608e1cee6c061828171c118 \
- --hash=sha256:6675329885c44657f826ef01d9e4fb33b9158e9d93c537d84ad8399539bc6f69 \
- --hash=sha256:75c1a6dfac6abd407634420c93864a1e274ebc1c7531346d9254c0d8f6ca00f9 \
- --hash=sha256:75da8f28eff26defba42c52986de97b22106cb8f26515b7c22443ebc9c2d3261 \
- --hash=sha256:77efb033d8d7ff233385f30c62c7c79271c8885d5c9657d967ede124671bbdfb \
- --hash=sha256:78a7d3ab09dc47ac1a363a493e6112d8cabed7ba7caad5f54dbe2f08676d1b47 \
- --hash=sha256:7c7db70d57e5e1089a274cbb2b1fd635c9a24de809a231b154965d415d6c6d24 \
- --hash=sha256:8c56c488ab471628ff3bfa80964372fc13504ece601e0d97a78ee74126b2045c \
- --hash=sha256:91669ccac46bbc1d09e9273546181919064e8df73488ea087dcac3e2968df9ba \
- --hash=sha256:9b666a475a65f4e839d3d10473fad6d47e0a9db14a2f4a224029c5bfde58ad2c \
- --hash=sha256:9cfef3ab326780c04d6646f68d4b4742aae222e8b8ea1d627c74e38afcbc9d91 \
- --hash=sha256:a13fc8aeb24bad755eea8f7f9d409438eb94e82cf86b08fe77a03fbc8f6a96b1 \
- --hash=sha256:a75c301f96db737e1c5ed5fd7d77d9c34466de16095a266509e13da09751bd19 \
- --hash=sha256:a76d4cb80f41ba94a6691264be76435e5f72f2cb3cab0b092a6212855f71c2f6 \
- --hash=sha256:aed04cabe26f30c1647ef0e8fbb207516fd40fe9472e9439695f5c6998e60ac5 \
- --hash=sha256:b148b56f5de675ee16d45e769e69f87623a4944f7443850bf9a9376e628a89d2 \
- --hash=sha256:b501c862d4901792adaec7c25b1ecc749e2662543f68bb194c42ba18d6eec98d \
- --hash=sha256:b846a1fcf8beadeb9ea4f44ec5bdde393e2f1569e17d700bfc49cd69bde75881 \
- --hash=sha256:b931ae8f62db78861b0ff1ac017851764602288575d65b8e8ff1963fed419063 \
- --hash=sha256:c33ab3ca9d3ccd581d58e989d67554e42d8d4ded94ab3ade3508455fe70e65f7 \
- --hash=sha256:c6604b735bb12fef8e0efd5578c9fb5d3d8532d5001ea13a19cddf295673ee09 \
- --hash=sha256:d8db08051fc9e7d8bc622f2112511b8107d8f27cd89e2f64ec45e9825e8288da \
- --hash=sha256:d9203500f7c63545b4ce3799319fe4d9feb1a1b89b28d3cb5abd11b9dd64147e \
- --hash=sha256:dc492779501fa723b04d0ab1f5be046797fee17d27700476edc7ee9ae535a61e \
- --hash=sha256:e6bcdf33aec38d16508ce61fd81838f24c83c90a1d1b8c68982857038673d6b8 \
- --hash=sha256:e76ce097e3c57c4bcb67c5aa24a0ecdbd9f74ea9219997a707a4061fbe2707aa \
- --hash=sha256:eff1ac3cc66c2ac7cda1e64b4e2f3ffef474b7335f92fc3833fc632d595fcee6 \
- --hash=sha256:f3cb4a569029b9f291f88aafc927dd53683757e640081ca8c412781ea144565e \
- --hash=sha256:f79b168428351d11e10c5aeb61a74e1851ec221081299f4cf56036a95431c43a \
- --hash=sha256:fa646ecec9528bef693415c79a86e733c70a4965dd938e9a226b0fc64c9d2e6c \
- --hash=sha256:fe2efccb324948a11dd09d22136fe2ac8a97d6c1347cf0b58a911dcd529f66b7 \
- --hash=sha256:fff4f534200a04b4a36e7ae3cb74493afe807b517a09e99cb4faa89a34ed6ecd
-fsspec==2026.2.0 \
- --hash=sha256:6544e34b16869f5aacd5b90bdf1a71acb37792ea3ddf6125ee69a22a53fb8bff \
- --hash=sha256:98de475b5cb3bd66bedd5c4679e87b4fdfe1a3bf4d707b151b3c07e58c9a2437
-graphemeu==0.7.2 \
- --hash=sha256:1444520f6899fd30114fc2a39f297d86d10fa0f23bf7579f772f8bc7efaa2542 \
- --hash=sha256:42bbe373d7c146160f286cd5f76b1a8ad29172d7333ce10705c5cc282462a4f8
-jinja2==3.1.6 \
- --hash=sha256:85ece4451f492d0c13c5dd7c13a64681a86afae63a5f347908daf103ce6d2f67
-joblib==1.5.3 \
- --hash=sha256:5fc3c5039fc5ca8c0276333a188bbd59d6b7ab37fe6632daa76bc7f9ec18e713 \
- --hash=sha256:8561a3269e6801106863fd0d6d84bb737be9e7631e33aaed3fb9ce5953688da3
-jsonschema==4.26.0 \
- --hash=sha256:0c26707e2efad8aa1bfc5b7ce170f3fccc2e4918ff85989ba9ffa9facb2be326 \
- --hash=sha256:d489f15263b8d200f8387e64b4c3a75f06629559fb73deb8fdfb525f2dab50ce
-jsonschema-specifications==2025.9.1 \
- --hash=sha256:98802fee3a11ee76ecaca44429fda8a41bff98b00a0f2838151b113f210cc6fe \
- --hash=sha256:b540987f239e745613c7a9176f3edb72b832a4ac465cf02712288397832b5e8d
-kiwisolver==1.4.9 \
- --hash=sha256:0749fd8f4218ad2e851e11cc4dc05c7cbc0cbc4267bdfdb31782e65aace4ee9c \
- --hash=sha256:0763515d4df10edf6d06a3c19734e2566368980d21ebec439f33f9eb936c07b7 \
- --hash=sha256:0856e241c2d3df4efef7c04a1e46b1936b6120c9bcf36dd216e3acd84bc4fb21 \
- --hash=sha256:0a590506f303f512dff6b7f75fd2fd18e16943efee932008fe7140e5fa91d80e \
- --hash=sha256:0ab74e19f6a2b027ea4f845a78827969af45ce790e6cb3e1ebab71bdf9f215ff \
- --hash=sha256:0ae37737256ba2de764ddc12aed4956460277f00c4996d51a197e72f62f5eec7 \
- --hash=sha256:0e4e2bf29574a6a7b7f6cb5fa69293b9f96c928949ac4a53ba3f525dffb87f9c \
- --hash=sha256:15163165efc2f627eb9687ea5f3a28137217d217ac4024893d753f46bce9de26 \
- --hash=sha256:17680d737d5335b552994a2008fab4c851bcd7de33094a82067ef3a576ff02fa \
- --hash=sha256:1a12cf6398e8a0a001a059747a1cbf24705e18fe413bc22de7b3d15c67cffe3f \
- --hash=sha256:1b11d6a633e4ed84fc0ddafd4ebfd8ea49b3f25082c04ad12b8315c11d504dc1 \
- --hash=sha256:1fa333e8b2ce4d9660f2cda9c0e1b6bafcfb2457a9d259faa82289e73ec24891 \
- --hash=sha256:2327a4a30d3ee07d2fbe2e7933e8a37c591663b96ce42a00bc67461a87d7df77 \
- --hash=sha256:2405a7d98604b87f3fc28b1716783534b1b4b8510d8142adca34ee0bc3c87543 \
- --hash=sha256:2489e4e5d7ef9a1c300a5e0196e43d9c739f066ef23270607d45aba368b91f2d \
- --hash=sha256:24c175051354f4a28c5d6a31c93906dc653e2bf234e8a4bbfb964892078898ce \
- --hash=sha256:2635d352d67458b66fd0667c14cb1d4145e9560d503219034a18a87e971ce4f3 \
- --hash=sha256:2c1a4f57df73965f3f14df20b80ee29e6a7930a57d2d9e8491a25f676e197c60 \
- --hash=sha256:2c93f00dcba2eea70af2be5f11a830a742fe6b579a1d4e00f47760ef13be247a \
- --hash=sha256:39a219e1c81ae3b103643d2aedb90f1ef22650deb266ff12a19e7773f3e5f089 \
- --hash=sha256:3b3115b2581ea35bb6d1f24a4c90af37e5d9b49dcff267eeed14c3893c5b86ab \
- --hash=sha256:40092754720b174e6ccf9e845d0d8c7d8e12c3d71e7fc35f55f3813e96376f78 \
- --hash=sha256:412f287c55a6f54b0650bd9b6dce5aceddb95864a1a90c87af16979d37c89771 \
- --hash=sha256:464415881e4801295659462c49461a24fb107c140de781d55518c4b80cb6790f \
- --hash=sha256:497d05f29a1300d14e02e6441cf0f5ee81c1ff5a304b0d9fb77423974684e08b \
- --hash=sha256:4a2899935e724dd1074cb568ce7ac0dce28b2cd6ab539c8e001a8578eb106d14 \
- --hash=sha256:4a48a2ce79d65d363597ef7b567ce3d14d68783d2b2263d98db3d9477805ba32 \
- --hash=sha256:4d1d9e582ad4d63062d34077a9a1e9f3c34088a2ec5135b1f7190c07cf366527 \
- --hash=sha256:52a15b0f35dad39862d376df10c5230155243a2c1a436e39eb55623ccbd68185 \
- --hash=sha256:540c7c72324d864406a009d72f5d6856f49693db95d1fbb46cf86febef873634 \
- --hash=sha256:5656aa670507437af0207645273ccdfee4f14bacd7f7c67a4306d0dcaeaf6eed \
- --hash=sha256:5a0f2724dfd4e3b3ac5a82436a8e6fd16baa7d507117e4279b660fe8ca38a3a1 \
- --hash=sha256:60c439763a969a6af93b4881db0eed8fadf93ee98e18cbc35bc8da868d0c4f0c \
- --hash=sha256:61874cdb0a36016354853593cffc38e56fc9ca5aa97d2c05d3dcf6922cd55a11 \
- --hash=sha256:67bb8b474b4181770f926f7b7d2f8c0248cbcb78b660fdd41a47054b28d2a752 \
- --hash=sha256:720e05574713db64c356e86732c0f3c5252818d05f9df320f0ad8380641acea5 \
- --hash=sha256:72d0eb9fba308b8311685c2268cf7d0a0639a6cd027d8128659f72bdd8a024b4 \
- --hash=sha256:767c23ad1c58c9e827b649a9ab7809fd5fd9db266a9cf02b0e926ddc2c680d58 \
- --hash=sha256:77937e5e2a38a7b48eef0585114fe7930346993a88060d0bf886086d2aa49ef5 \
- --hash=sha256:7a08b491ec91b1d5053ac177afe5290adacf1f0f6307d771ccac5de30592d198 \
- --hash=sha256:7b4da0d01ac866a57dd61ac258c5607b4cd677f63abaec7b148354d2b2cdd536 \
- --hash=sha256:7cf974dd4e35fa315563ac99d6287a1024e4dc2077b8a7d7cd3d2fb65d283134 \
- --hash=sha256:84fd60810829c27ae375114cd379da1fa65e6918e1da405f356a775d49a62bcf \
- --hash=sha256:858e4c22fb075920b96a291928cb7dea5644e94c0ee4fcd5af7e865655e4ccf2 \
- --hash=sha256:85b5352f94e490c028926ea567fc569c52ec79ce131dadb968d3853e809518c2 \
- --hash=sha256:85bd218b5ecfbee8c8a82e121802dcb519a86044c9c3b2e4aef02fa05c6da370 \
- --hash=sha256:8a1f570ce4d62d718dce3f179ee78dac3b545ac16c0c04bb363b7607a949c0d1 \
- --hash=sha256:8fdca1def57a2e88ef339de1737a1449d6dbf5fab184c54a1fca01d541317154 \
- --hash=sha256:90f47e70293fc3688b71271100a1a5453aa9944a81d27ff779c108372cf5567b \
- --hash=sha256:92a2f997387a1b79a75e7803aa7ded2cfbe2823852ccf1ba3bcf613b62ae3197 \
- --hash=sha256:9928fe1eb816d11ae170885a74d074f57af3a0d65777ca47e9aeb854a1fba386 \
- --hash=sha256:9af39d6551f97d31a4deebeac6f45b156f9755ddc59c07b402c148f5dbb6482a \
- --hash=sha256:9cf554f21be770f5111a1690d42313e140355e687e05cf82cb23d0a721a64a48 \
- --hash=sha256:a30fd6fdef1430fd9e1ba7b3398b5ee4e2887783917a687d86ba69985fb08748 \
- --hash=sha256:a31d512c812daea6d8b3be3b2bfcbeb091dbb09177706569bcfc6240dcf8b41c \
- --hash=sha256:a5d0432ccf1c7ab14f9949eec60c5d1f924f17c037e9f8b33352fa05799359b8 \
- --hash=sha256:a60ea74330b91bd22a29638940d115df9dc00af5035a9a2a6ad9399ffb4ceca5 \
- --hash=sha256:ac5a486ac389dddcc5bef4f365b6ae3ffff2c433324fb38dd35e3fab7c957999 \
- --hash=sha256:aedff62918805fb62d43a4aa2ecd4482c380dc76cd31bd7c8878588a61bd0369 \
- --hash=sha256:b34e51affded8faee0dfdb705416153819d8ea9250bbbf7ea1b249bdeb5f1122 \
- --hash=sha256:b4b4d74bda2b8ebf4da5bd42af11d02d04428b2c32846e4c2c93219df8a7987b \
- --hash=sha256:b67e6efbf68e077dd71d1a6b37e43e1a99d0bff1a3d51867d45ee8908b931098 \
- --hash=sha256:b78efa4c6e804ecdf727e580dbb9cba85624d2e1c6b5cb059c66290063bd99a9 \
- --hash=sha256:bb4ae2b57fc1d8cbd1cf7b1d9913803681ffa903e7488012be5b76dedf49297f \
- --hash=sha256:bdd1a81a1860476eb41ac4bc1e07b3f07259e6d55bbf739b79c8aaedcf512799 \
- --hash=sha256:bdee92c56a71d2b24c33a7d4c2856bd6419d017e08caa7802d2963870e315028 \
- --hash=sha256:be6a04e6c79819c9a8c2373317d19a96048e5a3f90bec587787e86a1153883c2 \
- --hash=sha256:bfc08add558155345129c7803b3671cf195e6a56e7a12f3dde7c57d9b417f525 \
- --hash=sha256:c3b22c26c6fd6811b0ae8363b95ca8ce4ea3c202d3d0975b2914310ceb1bcc4d \
- --hash=sha256:c9e7cdf45d594ee04d5be1b24dd9d49f3d1590959b2271fb30b5ca2b262c00fb \
- --hash=sha256:cb27e7b78d716c591e88e0a09a2139c6577865d7f2e152488c2cc6257f460872 \
- --hash=sha256:cc9617b46837c6468197b5945e196ee9ca43057bb7d9d1ae688101e4e1dddf64 \
- --hash=sha256:ccd09f20ccdbbd341b21a67ab50a119b64a403b09288c27481575105283c1586 \
- --hash=sha256:ce6a3a4e106cf35c2d9c4fa17c05ce0b180db622736845d4315519397a77beaf \
- --hash=sha256:d0005b053977e7b43388ddec89fa567f43d4f6d5c2c0affe57de5ebf290dc552 \
- --hash=sha256:d4188e73af84ca82468f09cadc5ac4db578109e52acb4518d8154698d3a87ca2 \
- --hash=sha256:d4efec7bcf21671db6a3294ff301d2fc861c31faa3c8740d1a94689234d1b415 \
- --hash=sha256:d75aa530ccfaa593da12834b86a0724f58bff12706659baa9227c2ccaa06264c \
- --hash=sha256:d84cd4061ae292d8ac367b2c3fa3aad11cb8625a95d135fe93f286f914f3f5a6 \
- --hash=sha256:d8aacd3d4b33b772542b2e01beb50187536967b514b00003bdda7589722d2a64 \
- --hash=sha256:d8fc5c867c22b828001b6a38d2eaeb88160bf5783c6cb4a5e440efc981ce286d \
- --hash=sha256:d976bbb382b202f71c67f77b0ac11244021cfa3f7dfd9e562eefcea2df711548 \
- --hash=sha256:dba5ee5d3981160c28d5490f0d1b7ed730c22470ff7f6cc26cfcfaacb9896a07 \
- --hash=sha256:dc1ae486f9abcef254b5618dfb4113dd49f94c68e3e027d03cf0143f3f772b61 \
- --hash=sha256:dd0a578400839256df88c16abddf9ba14813ec5f21362e1fe65022e00c883d4d \
- --hash=sha256:deed0c7258ceb4c44ad5ec7d9918f9f14fd05b2be86378d86cf50e63d1e7b771 \
- --hash=sha256:e09c2279a4d01f099f52d5c4b3d9e208e91edcbd1a175c9662a8b16e000fece9 \
- --hash=sha256:e2ea9f7ab7fbf18fffb1b5434ce7c69a07582f7acc7717720f1d69f3e806f90c \
- --hash=sha256:e6b93f13371d341afee3be9f7c5964e3fe61d5fa30f6a30eb49856935dfe4fc3 \
- --hash=sha256:eb14a5da6dc7642b0f3a18f13654847cd8b7a2550e2645a5bda677862b03ba16 \
- --hash=sha256:ed0fecd28cc62c54b262e3736f8bb2512d8dcfdc2bcf08be5f47f96bf405b145 \
- --hash=sha256:ede8c6d533bc6601a47ad4046080d36b8fc99f81e6f1c17b0ac3c2dc91ac7611 \
- --hash=sha256:efb3a45b35622bb6c16dbfab491a8f5a391fe0e9d45ef32f4df85658232ca0e2 \
- --hash=sha256:f117e1a089d9411663a3207ba874f31be9ac8eaa5b533787024dc07aeb74f464 \
- --hash=sha256:f2ba92255faa7309d06fe44c3a4a97efe1c8d640c2a79a5ef728b685762a6fd2 \
- --hash=sha256:f6008a4919fdbc0b0097089f67a1eb55d950ed7e90ce2cc3e640abadd2757a04 \
- --hash=sha256:f68208a520c3d86ea51acf688a3e3002615a7f0238002cccc17affecc86a8a54 \
- --hash=sha256:f68e4f3eeca8fb22cc3d731f9715a13b652795ef657a13df1ad0c7dc0e9731df \
- --hash=sha256:fb3b8132019ea572f4611d770991000d7f58127560c4889729248eb5852a102f \
- --hash=sha256:fb940820c63a9590d31d88b815e7a3aa5915cad3ce735ab45f0c730b39547de1 \
- --hash=sha256:fc1795ac5cd0510207482c3d1d3ed781143383b8cfd36f5c645f3897ce066220
-markdown-it-py==4.0.0 \
- --hash=sha256:87327c59b172c5011896038353a81343b6754500a08cd7a4973bb48c6d578147 \
- --hash=sha256:cb0a2b4aa34f932c007117b194e945bd74e0ec24133ceb5bac59009cda1cb9f3
-markupsafe==3.0.3 \
- --hash=sha256:f42d0984e947b8adf7dd6dde396e720934d12c506ce84eea8476409563607591
-matplotlib==3.10.8 \
- --hash=sha256:00270d217d6b20d14b584c521f810d60c5c78406dc289859776550df837dcda7 \
- --hash=sha256:0a33deb84c15ede243aead39f77e990469fff93ad1521163305095b77b72ce4a \
- --hash=sha256:113bb52413ea508ce954a02c10ffd0d565f9c3bc7f2eddc27dfe1731e71c7b5f \
- --hash=sha256:12d90df9183093fcd479f4172ac26b322b1248b15729cb57f42f71f24c7e37a3 \
- --hash=sha256:15d30132718972c2c074cd14638c7f4592bd98719e2308bccea40e0538bc0cb5 \
- --hash=sha256:18821ace09c763ec93aef5eeff087ee493a24051936d7b9ebcad9662f66501f9 \
- --hash=sha256:1ae029229a57cd1e8fe542485f27e7ca7b23aa9e8944ddb4985d0bc444f1eca2 \
- --hash=sha256:2299372c19d56bcd35cf05a2738308758d32b9eaed2371898d8f5bd33f084aa3 \
- --hash=sha256:238b7ce5717600615c895050239ec955d91f321c209dd110db988500558e70d6 \
- --hash=sha256:24d50994d8c5816ddc35411e50a86ab05f575e2530c02752e02538122613371f \
- --hash=sha256:25d380fe8b1dc32cf8f0b1b448470a77afb195438bafdf1d858bfb876f3edf7b \
- --hash=sha256:2c1998e92cd5999e295a731bcb2911c75f597d937341f3030cc24ef2733d78a8 \
- --hash=sha256:2cf5bd12cecf46908f286d7838b2abc6c91cda506c0445b8223a7c19a00df008 \
- --hash=sha256:32f8dce744be5569bebe789e46727946041199030db8aeb2954d26013a0eb26b \
- --hash=sha256:37b3c1cc42aa184b3f738cfa18c1c1d72fd496d85467a6cf7b807936d39aa656 \
- --hash=sha256:3a48a78d2786784cc2413e57397981fb45c79e968d99656706018d6e62e57958 \
- --hash=sha256:3ab4aabc72de4ff77b3ec33a6d78a68227bf1123465887f9905ba79184a1cc04 \
- --hash=sha256:3c624e43ed56313651bc18a47f838b60d7b8032ed348911c54906b130b20071b \
- --hash=sha256:3f2e409836d7f5ac2f1c013110a4d50b9f7edc26328c108915f9075d7d7a91b6 \
- --hash=sha256:3f5c3e4da343bba819f0234186b9004faba952cc420fbc522dc4e103c1985908 \
- --hash=sha256:41703cc95688f2516b480f7f339d8851a6035f18e100ee6a32bc0b8536a12a9c \
- --hash=sha256:495672de149445ec1b772ff2c9ede9b769e3cb4f0d0aa7fa730d7f59e2d4e1c1 \
- --hash=sha256:4cf267add95b1c88300d96ca837833d4112756045364f5c734a2276038dae27d \
- --hash=sha256:56271f3dac49a88d7fca5060f004d9d22b865f743a12a23b1e937a0be4818ee1 \
- --hash=sha256:595ba4d8fe983b88f0eec8c26a241e16d6376fe1979086232f481f8f3f67494c \
- --hash=sha256:5f62550b9a30afde8c1c3ae450e5eb547d579dd69b25c2fc7a1c67f934c1717a \
- --hash=sha256:646d95230efb9ca614a7a594d4fcacde0ac61d25e37dd51710b36477594963ce \
- --hash=sha256:64fcc24778ca0404ce0cb7b6b77ae1f4c7231cdd60e6778f999ee05cbd581b9a \
- --hash=sha256:6be43b667360fef5c754dda5d25a32e6307a03c204f3c0fc5468b78fa87b4160 \
- --hash=sha256:6da7c2ce169267d0d066adcf63758f0604aa6c3eebf67458930f9d9b79ad1db1 \
- --hash=sha256:83d282364ea9f3e52363da262ce32a09dfe241e4080dcedda3c0db059d3c1f11 \
- --hash=sha256:9153c3292705be9f9c64498a8872118540c3f4123d1a1c840172edf262c8be4a \
- --hash=sha256:99eefd13c0dc3b3c1b4d561c1169e65fe47aab7b8158754d7c084088e2329466 \
- --hash=sha256:a0a7f52498f72f13d4a25ea70f35f4cb60642b466cbb0a9be951b5bc3f45a486 \
- --hash=sha256:a2b336e2d91a3d7006864e0990c83b216fcdca64b5a6484912902cef87313d78 \
- --hash=sha256:a48f2b74020919552ea25d222d5cc6af9ca3f4eb43a93e14d068457f545c2a17 \
- --hash=sha256:ad3d9833a64cf48cc4300f2b406c3d0f4f4724a91c0bd5640678a6ba7c102077 \
- --hash=sha256:b44d07310e404ba95f8c25aa5536f154c0a8ec473303535949e52eb71d0a1565 \
- --hash=sha256:b53285e65d4fa4c86399979e956235deb900be5baa7fc1218ea67fbfaeaadd6f \
- --hash=sha256:b5a2b97dbdc7d4f353ebf343744f1d1f1cca8aa8bfddb4262fcf4306c3761d50 \
- --hash=sha256:b9a5ca4ac220a0cdd1ba6bcba3608547117d30468fefce49bb26f55c1a3d5c58 \
- --hash=sha256:bab485bcf8b1c7d2060b4fcb6fc368a9e6f4cd754c9c2fea281f4be21df394a2 \
- --hash=sha256:c108a1d6fa78a50646029cb6d49808ff0fc1330fda87fa6f6250c6b5369b6645 \
- --hash=sha256:d56a1efd5bfd61486c8bc968fa18734464556f0fb8e51690f4ac25d85cbbbbc2 \
- --hash=sha256:d9050fee89a89ed57b4fb2c1bfac9a3d0c57a0d55aed95949eedbc42070fea39 \
- --hash=sha256:dd80ecb295460a5d9d260df63c43f4afbdd832d725a531f008dad1664f458adf \
- --hash=sha256:e8ea3e2d4066083e264e75c829078f9e149fa119d27e19acd503de65e0b13149 \
- --hash=sha256:eb3823f11823deade26ce3b9f40dcb4a213da7a670013929f31d5f5ed1055b22 \
- --hash=sha256:ee40c27c795bda6a5292e9cff9890189d32f7e3a0bf04e0e3c9430c4a00c37df \
- --hash=sha256:efb30e3baaea72ce5928e32bab719ab4770099079d66726a62b11b1ef7273be4 \
- --hash=sha256:f254d118d14a7f99d616271d6c3c27922c092dac11112670b157798b89bf4933 \
- --hash=sha256:f89c151aab2e2e23cb3fe0acad1e8b82841fd265379c4cecd0f3fcb34c15e0f6 \
- --hash=sha256:f97aeb209c3d2511443f8797e3e5a569aebb040d4f8bc79aa3ee78a8fb9e3dd8 \
- --hash=sha256:f9b587c9c7274c1613a30afabf65a272114cd6cdbe67b3406f818c79d7ab2e2a \
- --hash=sha256:fb061f596dad3a0f52b60dc6a5dec4a0c300dec41e058a7efe09256188d170b7
-mdurl==0.1.2 \
- --hash=sha256:84008a41e51615a49fc9966191ff91509e3c40b939176e643fd50a5c2196b8f8 \
- --hash=sha256:bb413d29f5eea38f31dd4754dd7377d4465116fb207585f97bf925588687c1ba
-moocore==0.2.0 \
- --hash=sha256:3dc601f85f9a4743ed50ddd027dca30e3bb55c899916a092c2ece495b1b2de08 \
- --hash=sha256:653449231f328d3c9e69693ec3d44e8c77f38ab7e9ef0c69dd9ded40449e980d \
- --hash=sha256:a7683feddfd2a47b4a0f89ee8d370cae72331792f68e67f083ccb37bb2f1c8cf \
- --hash=sha256:b90c7bde2164f9b95c6b2e870f0ca6ccc5dabff2bf8086162d7318c770e5868f \
- --hash=sha256:cf8f091a7304532ed605acd82acd051e89af22ece8e2a27a3cee0faf9f2ea185 \
- --hash=sha256:d0699b770b5eebdeac477a356d539efa8807c6cf067a453a0682e0df2299a512 \
- --hash=sha256:e93c07062adefd0fcba73a521f325f7fb874f2af92aaeec203cf9db31a41894b \
- --hash=sha256:ea057409731e73dbc4ba4214cbf7747309695b01314f8786678b758cc9c561c4
-mpmath==1.3.0 \
- --hash=sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f \
- --hash=sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c
-natsort==8.4.0 \
- --hash=sha256:45312c4a0e5507593da193dedd04abb1469253b601ecaf63445ad80f0a1ea581 \
- --hash=sha256:4732914fb471f56b5cce04d7bae6f164a592c7712e1c85f9ef585e197299521c
-networkx==3.4.2 \
- --hash=sha256:307c3669428c5362aab27c8a1260aa8f47c4e91d3891f48be0141738d8d053e1 \
- --hash=sha256:df5d4365b724cf81b8c6a7312509d0c22386097011ad1abe274afd5e9d3bbc5f
-ninja==1.13.0 \
- --hash=sha256:11be2d22027bde06f14c343f01d31446747dbb51e72d00decca2eb99be911e2f \
- --hash=sha256:1c97223cdda0417f414bf864cfb73b72d8777e57ebb279c5f6de368de0062988 \
- --hash=sha256:3c0b40b1f0bba764644385319028650087b4c1b18cdfa6f45cb39a3669b81aa9 \
- --hash=sha256:3d00c692fb717fd511abeb44b8c5d00340c36938c12d6538ba989fe764e79630 \
- --hash=sha256:3d7d7779d12cb20c6d054c61b702139fd23a7a964ec8f2c823f1ab1b084150db \
- --hash=sha256:4a40ce995ded54d9dc24f8ea37ff3bf62ad192b547f6c7126e7e25045e76f978 \
- --hash=sha256:4be9c1b082d244b1ad7ef41eb8ab088aae8c109a9f3f0b3e56a252d3e00f42c1 \
- --hash=sha256:5f8e1e8a1a30835eeb51db05cf5a67151ad37542f5a4af2a438e9490915e5b72 \
- --hash=sha256:60056592cf495e9a6a4bea3cd178903056ecb0943e4de45a2ea825edb6dc8d3e \
- --hash=sha256:6739d3352073341ad284246f81339a384eec091d9851a886dfa5b00a6d48b3e2 \
- --hash=sha256:8cfbb80b4a53456ae8a39f90ae3d7a2129f45ea164f43fadfa15dc38c4aef1c9 \
- --hash=sha256:aa45b4037b313c2f698bc13306239b8b93b4680eb47e287773156ac9e9304714 \
- --hash=sha256:b4f2a072db3c0f944c32793e91532d8948d20d9ab83da9c0c7c15b5768072200 \
- --hash=sha256:be7f478ff9f96a128b599a964fc60a6a87b9fa332ee1bd44fa243ac88d50291c \
- --hash=sha256:d741a5e6754e0bda767e3274a0f0deeef4807f1fec6c0d7921a0244018926ae5 \
- --hash=sha256:e8bad11f8a00b64137e9b315b137d8bb6cbf3086fbdc43bf1f90fd33324d2e96 \
- --hash=sha256:fa2a8bfc62e31b08f83127d1613d10821775a0eb334197154c4d6067b7068ff1 \
- --hash=sha256:fb46acf6b93b8dd0322adc3a4945452a4e774b75b91293bafcc7b7f8e6517dfa \
- --hash=sha256:fb8ee8719f8af47fed145cced4a85f0755dd55d45b2bddaf7431fa89803c5f3e
-nncf==2.19.0 \
- --hash=sha256:6c782b0e5a8120d36a3249cd61edaf311b6270dd90095661700658899726e523 \
- --hash=sha256:d32a9667e6b2837557bdd340895bb49e8d103eb1ec004031386e8c2bb4554cac
-numpy==2.2.6 \
- --hash=sha256:038613e9fb8c72b0a41f025a7e4c3f0b7a1b5d768ece4796b674c8f3fe13efff \
- --hash=sha256:0678000bb9ac1475cd454c6b8c799206af8107e310843532b04d49649c717a47 \
- --hash=sha256:0811bb762109d9708cca4d0b13c4f67146e3c3b7cf8d34018c722adb2d957c84 \
- --hash=sha256:0b605b275d7bd0c640cad4e5d30fa701a8d59302e127e5f79138ad62762c3e3d \
- --hash=sha256:0bca768cd85ae743b2affdc762d617eddf3bcf8724435498a1e80132d04879e6 \
- --hash=sha256:1bc23a79bfabc5d056d106f9befb8d50c31ced2fbc70eedb8155aec74a45798f \
- --hash=sha256:287cc3162b6f01463ccd86be154f284d0893d2b3ed7292439ea97eafa8170e0b \
- --hash=sha256:37c0ca431f82cd5fa716eca9506aefcabc247fb27ba69c5062a6d3ade8cf8f49 \
- --hash=sha256:37e990a01ae6ec7fe7fa1c26c55ecb672dd98b19c3d0e1d1f326fa13cb38d163 \
- --hash=sha256:389d771b1623ec92636b0786bc4ae56abafad4a4c513d36a55dce14bd9ce8571 \
- --hash=sha256:3d70692235e759f260c3d837193090014aebdf026dfd167834bcba43e30c2a42 \
- --hash=sha256:41c5a21f4a04fa86436124d388f6ed60a9343a6f767fced1a8a71c3fbca038ff \
- --hash=sha256:481b49095335f8eed42e39e8041327c05b0f6f4780488f61286ed3c01368d491 \
- --hash=sha256:4eeaae00d789f66c7a25ac5f34b71a7035bb474e679f410e5e1a94deb24cf2d4 \
- --hash=sha256:55a4d33fa519660d69614a9fad433be87e5252f4b03850642f88993f7b2ca566 \
- --hash=sha256:5a6429d4be8ca66d889b7cf70f536a397dc45ba6faeb5f8c5427935d9592e9cf \
- --hash=sha256:5bd4fc3ac8926b3819797a7c0e2631eb889b4118a9898c84f585a54d475b7e40 \
- --hash=sha256:5beb72339d9d4fa36522fc63802f469b13cdbe4fdab4a288f0c441b74272ebfd \
- --hash=sha256:6031dd6dfecc0cf9f668681a37648373bddd6421fff6c66ec1624eed0180ee06 \
- --hash=sha256:71594f7c51a18e728451bb50cc60a3ce4e6538822731b2933209a1f3614e9282 \
- --hash=sha256:74d4531beb257d2c3f4b261bfb0fc09e0f9ebb8842d82a7b4209415896adc680 \
- --hash=sha256:7befc596a7dc9da8a337f79802ee8adb30a552a94f792b9c9d18c840055907db \
- --hash=sha256:894b3a42502226a1cac872f840030665f33326fc3dac8e57c607905773cdcde3 \
- --hash=sha256:8e41fd67c52b86603a91c1a505ebaef50b3314de0213461c7a6e99c9a3beff90 \
- --hash=sha256:8e9ace4a37db23421249ed236fdcdd457d671e25146786dfc96835cd951aa7c1 \
- --hash=sha256:8fc377d995680230e83241d8a96def29f204b5782f371c532579b4f20607a289 \
- --hash=sha256:9551a499bf125c1d4f9e250377c1ee2eddd02e01eac6644c080162c0c51778ab \
- --hash=sha256:b0544343a702fa80c95ad5d3d608ea3599dd54d4632df855e4c8d24eb6ecfa1c \
- --hash=sha256:b093dd74e50a8cba3e873868d9e93a85b78e0daf2e98c6797566ad8044e8363d \
- --hash=sha256:b412caa66f72040e6d268491a59f2c43bf03eb6c96dd8f0307829feb7fa2b6fb \
- --hash=sha256:b4f13750ce79751586ae2eb824ba7e1e8dba64784086c98cdbbcc6a42112ce0d \
- --hash=sha256:b64d8d4d17135e00c8e346e0a738deb17e754230d7e0810ac5012750bbd85a5a \
- --hash=sha256:ba10f8411898fc418a521833e014a77d3ca01c15b0c6cdcce6a0d2897e6dbbdf \
- --hash=sha256:bd48227a919f1bafbdda0583705e547892342c26fb127219d60a5c36882609d1 \
- --hash=sha256:c1f9540be57940698ed329904db803cf7a402f3fc200bfe599334c9bd84a40b2 \
- --hash=sha256:c820a93b0255bc360f53eca31a0e676fd1101f673dda8da93454a12e23fc5f7a \
- --hash=sha256:ce47521a4754c8f4593837384bd3424880629f718d87c5d44f8ed763edd63543 \
- --hash=sha256:d042d24c90c41b54fd506da306759e06e568864df8ec17ccc17e9e884634fd00 \
- --hash=sha256:de749064336d37e340f640b05f24e9e3dd678c57318c7289d222a8a2f543e90c \
- --hash=sha256:e1dda9c7e08dc141e0247a5b8f49cf05984955246a327d4c48bda16821947b2f \
- --hash=sha256:e29554e2bef54a90aa5cc07da6ce955accb83f21ab5de01a62c8478897b264fd \
- --hash=sha256:e3143e4451880bed956e706a3220b4e5cf6172ef05fcc397f6f36a550b1dd868 \
- --hash=sha256:e8213002e427c69c45a52bbd94163084025f533a55a59d6f9c5b820774ef3303 \
- --hash=sha256:efd28d4e9cd7d7a8d39074a4d44c63eda73401580c5c76acda2ce969e0a38e83 \
- --hash=sha256:f0fd6321b839904e15c46e0d257fdd101dd7f530fe03fd6359c1ea63738703f3 \
- --hash=sha256:f1372f041402e37e5e633e586f62aa53de2eac8d98cbfb822806ce4bbefcb74d \
- --hash=sha256:f2618db89be1b4e05f7a1a847a9c1c0abd63e63a1607d892dd54668dd92faf87 \
- --hash=sha256:f447e6acb680fd307f40d3da4852208af94afdfab89cf850986c3ca00562f4fa \
- --hash=sha256:f92729c95468a2f4f15e9bb94c432a9229d0d50de67304399627a943201baa2f \
- --hash=sha256:f9f1adb22318e121c5c69a09142811a201ef17ab257a1e66ca3025065b7f53ae \
- --hash=sha256:fc0c5673685c508a142ca65209b4e79ed6740a4ed6b2267dbba90f34b0b3cfda \
- --hash=sha256:fc7b73d02efb0e18c000e9ad8b83480dfcd5dfd11065997ed4c6747470ae8915 \
- --hash=sha256:fd83c01228a688733f1ded5201c678f0c53ecc1006ffbc404db9f7a899ac6249 \
- --hash=sha256:fe27749d33bb772c80dcd84ae7e8df2adc920ae8297400dabec45f0dedb3f6de \
- --hash=sha256:fee4236c876c4e8369388054d02d0e9bb84821feb1a64dd59e137e6511a551f8
-openvino-telemetry==2025.2.0 \
- --hash=sha256:8bf8127218e51e99547bf38b8fb85a8b31c9bf96e6f3a82eb0b3b6a34155977c \
- --hash=sha256:bcb667e83a44f202ecf4cfa49281715c6d7e21499daec04ff853b7f964833599
-packaging==24.1 \
- --hash=sha256:5b8f2217dbdbd2f7f384c41c628544e6d52f2d0f53c6d0c3ea61aa5d1d7ff124
-pandas==2.3.3 \
- --hash=sha256:0242fe9a49aa8b4d78a4fa03acb397a58833ef6199e9aa40a95f027bb3a1b6e7 \
- --hash=sha256:1611aedd912e1ff81ff41c745822980c49ce4a7907537be8692c8dbc31924593 \
- --hash=sha256:1b07204a219b3b7350abaae088f451860223a52cfb8a6c53358e7948735158e5 \
- --hash=sha256:1d37b5848ba49824e5c30bedb9c830ab9b7751fd049bc7914533e01c65f79791 \
- --hash=sha256:23ebd657a4d38268c7dfbdf089fbc31ea709d82e4923c5ffd4fbd5747133ce73 \
- --hash=sha256:2462b1a365b6109d275250baaae7b760fd25c726aaca0054649286bcfbb3e8ec \
- --hash=sha256:28083c648d9a99a5dd035ec125d42439c6c1c525098c58af0fc38dd1a7a1b3d4 \
- --hash=sha256:2e3ebdb170b5ef78f19bfb71b0dc5dc58775032361fa188e814959b74d726dd5 \
- --hash=sha256:318d77e0e42a628c04dc56bcef4b40de67918f7041c2b061af1da41dcff670ac \
- --hash=sha256:371a4ab48e950033bcf52b6527eccb564f52dc826c02afd9a1bc0ab731bba084 \
- --hash=sha256:376c6446ae31770764215a6c937f72d917f214b43560603cd60da6408f183b6c \
- --hash=sha256:3869faf4bd07b3b66a9f462417d0ca3a9df29a9f6abd5d0d0dbab15dac7abe87 \
- --hash=sha256:3fd2f887589c7aa868e02632612ba39acb0b8948faf5cc58f0850e165bd46f35 \
- --hash=sha256:4793891684806ae50d1288c9bae9330293ab4e083ccd1c5e383c34549c6e4250 \
- --hash=sha256:4e0a175408804d566144e170d0476b15d78458795bb18f1304fb94160cabf40c \
- --hash=sha256:503cf027cf9940d2ceaa1a93cfb5f8c8c7e6e90720a2850378f0b3f3b1e06826 \
- --hash=sha256:5554c929ccc317d41a5e3d1234f3be588248e61f08a74dd17c9eabb535777dc9 \
- --hash=sha256:56851a737e3470de7fa88e6131f41281ed440d29a9268dcbf0002da5ac366713 \
- --hash=sha256:5caf26f64126b6c7aec964f74266f435afef1c1b13da3b0636c7518a1fa3e2b1 \
- --hash=sha256:602b8615ebcc4a0c1751e71840428ddebeb142ec02c786e8ad6b1ce3c8dec523 \
- --hash=sha256:6253c72c6a1d990a410bc7de641d34053364ef8bcd3126f7e7450125887dffe3 \
- --hash=sha256:6435cb949cb34ec11cc9860246ccb2fdc9ecd742c12d3304989017d53f039a78 \
- --hash=sha256:6d21f6d74eb1725c2efaa71a2bfc661a0689579b58e9c0ca58a739ff0b002b53 \
- --hash=sha256:6d2cefc361461662ac48810cb14365a365ce864afe85ef1f447ff5a1e99ea81c \
- --hash=sha256:74ecdf1d301e812db96a465a525952f4dde225fdb6d8e5a521d47e1f42041e21 \
- --hash=sha256:75ea25f9529fdec2d2e93a42c523962261e567d250b0013b16210e1d40d7c2e5 \
- --hash=sha256:854d00d556406bffe66a4c0802f334c9ad5a96b4f1f868adf036a21b11ef13ff \
- --hash=sha256:8fe25fc7b623b0ef6b5009149627e34d2a4657e880948ec3c840e9402e5c1b45 \
- --hash=sha256:900f47d8f20860de523a1ac881c4c36d65efcb2eb850e6948140fa781736e110 \
- --hash=sha256:93c2d9ab0fc11822b5eece72ec9587e172f63cff87c00b062f6e37448ced4493 \
- --hash=sha256:a16dcec078a01eeef8ee61bf64074b4e524a2a3f4b3be9326420cabe59c4778b \
- --hash=sha256:a21d830e78df0a515db2b3d2f5570610f5e6bd2e27749770e8bb7b524b89b450 \
- --hash=sha256:a45c765238e2ed7d7c608fc5bc4a6f88b642f2f01e70c0c23d2224dd21829d86 \
- --hash=sha256:a637c5cdfa04b6d6e2ecedcb81fc52ffb0fd78ce2ebccc9ea964df9f658de8c8 \
- --hash=sha256:a68e15f780eddf2b07d242e17a04aa187a7ee12b40b930bfdd78070556550e98 \
- --hash=sha256:b3d11d2fda7eb164ef27ffc14b4fcab16a80e1ce67e9f57e19ec0afaf715ba89 \
- --hash=sha256:b468d3dad6ff947df92dcb32ede5b7bd41a9b3cceef0a30ed925f6d01fb8fa66 \
- --hash=sha256:b98560e98cb334799c0b07ca7967ac361a47326e9b4e5a7dfb5ab2b1c9d35a1b \
- --hash=sha256:bdcd9d1167f4885211e401b3036c0c8d9e274eee67ea8d0758a256d60704cfe8 \
- --hash=sha256:bf1f8a81d04ca90e32a0aceb819d34dbd378a98bf923b6398b9a3ec0bf44de29 \
- --hash=sha256:c46467899aaa4da076d5abc11084634e2d197e9460643dd455ac3db5856b24d6 \
- --hash=sha256:c4fc4c21971a1a9f4bdb4c73978c7f7256caa3e62b323f70d6cb80db583350bc \
- --hash=sha256:c503ba5216814e295f40711470446bc3fd00f0faea8a086cbc688808e26f92a2 \
- --hash=sha256:d051c0e065b94b7a3cea50eb1ec32e912cd96dba41647eb24104b6c6c14c5788 \
- --hash=sha256:d3e28b3e83862ccf4d85ff19cf8c20b2ae7e503881711ff2d534dc8f761131aa \
- --hash=sha256:db4301b2d1f926ae677a751eb2bd0e8c5f5319c9cb3f88b0becbbb0b07b34151 \
- --hash=sha256:dd7478f1463441ae4ca7308a70e90b33470fa593429f9d4c578dd00d1fa78838 \
- --hash=sha256:e05e1af93b977f7eafa636d043f9f94c7ee3ac81af99c13508215942e64c993b \
- --hash=sha256:e19d192383eab2f4ceb30b412b22ea30690c9e618f78870357ae1d682912015a \
- --hash=sha256:e32e7cc9af0f1cc15548288a51a3b681cc2a219faa838e995f7dc53dbab1062d \
- --hash=sha256:ecaf1e12bdc03c86ad4a7ea848d66c685cb6851d807a26aa245ca3d2017a1908 \
- --hash=sha256:ee15f284898e7b246df8087fc82b87b01686f98ee67d85a17b7ab44143a3a9a0 \
- --hash=sha256:ee67acbbf05014ea6c763beb097e03cd629961c8a632075eeb34247120abcb4b \
- --hash=sha256:f086f6fe114e19d92014a1966f43a3e62285109afe874f067f5abbdcbb10e59c \
- --hash=sha256:f8bfc0e12dc78f777f323f55c58649591b2cd0c43534e8355c51d3fede5f4dee
-pillow==12.1.1 \
- --hash=sha256:02f84dfad02693676692746df05b89cf25597560db2857363a208e393429f5e9 \
- --hash=sha256:0330d233c1a0ead844fc097a7d16c0abff4c12e856c0b325f231820fee1f39da \
- --hash=sha256:03edcc34d688572014ff223c125a3f77fb08091e4607e7745002fc214070b35f \
- --hash=sha256:097690ba1f2efdeb165a20469d59d8bb03c55fb6621eb2041a060ae8ea3e9642 \
- --hash=sha256:178aa072084bd88ec759052feca8e56cbb14a60b39322b99a049e58090479713 \
- --hash=sha256:18e5bddd742a44b7e6b1e773ab5db102bd7a94c32555ba656e76d319d19c3850 \
- --hash=sha256:1a9b0ee305220b392e1124a764ee4265bd063e54a751a6b62eff69992f457fa9 \
- --hash=sha256:1f1625b72740fdda5d77b4def688eb8fd6490975d06b909fd19f13f391e077e0 \
- --hash=sha256:1f1be78ce9466a7ee64bfda57bdba0f7cc499d9794d518b854816c41bf0aa4e9 \
- --hash=sha256:1f90cff8aa76835cba5769f0b3121a22bd4eb9e6884cfe338216e557a9a548b8 \
- --hash=sha256:21329ec8c96c6e979cd0dfd29406c40c1d52521a90544463057d2aaa937d66a6 \
- --hash=sha256:2815a87ab27848db0321fb78c7f0b2c8649dee134b7f2b80c6a45c6831d75ccd \
- --hash=sha256:2c1fc0f2ca5f96a3c8407e41cca26a16e46b21060fe6d5b099d2cb01412222f5 \
- --hash=sha256:2e0c664be47252947d870ac0d327fea7e63985a08794758aa8af5b6cb6ec0c9c \
- --hash=sha256:339ffdcb7cbeaa08221cd401d517d4b1fe7a9ed5d400e4a8039719238620ca35 \
- --hash=sha256:344cf1e3dab3be4b1fa08e449323d98a2a3f819ad20f4b22e77a0ede31f0faa1 \
- --hash=sha256:36341d06738a9f66c8287cf8b876d24b18db9bd8740fa0672c74e259ad408cff \
- --hash=sha256:365b10bb9417dd4498c0e3b128018c4a624dc11c7b97d8cc54effe3b096f4c38 \
- --hash=sha256:3a5cbdcddad0af3da87cb16b60d23648bc3b51967eb07223e9fed77a82b457c4 \
- --hash=sha256:417423db963cb4be8bac3fc1204fe61610f6abeed1580a7a2cbb2fbda20f12af \
- --hash=sha256:42fc1f4677106188ad9a55562bbade416f8b55456f522430fadab3cef7cd4e60 \
- --hash=sha256:44ce27545b6efcf0fdbdceb31c9a5bdea9333e664cda58a7e674bb74608b3986 \
- --hash=sha256:472a8d7ded663e6162dafdf20015c486a7009483ca671cece7a9279b512fcb13 \
- --hash=sha256:47b94983da0c642de92ced1702c5b6c292a84bd3a8e1d1702ff923f183594717 \
- --hash=sha256:495c302af3aad1ca67420ddd5c7bd480c8867ad173528767d906428057a11f0e \
- --hash=sha256:4ceb838d4bd9dab43e06c363cab2eebf63846d6a4aeaea283bbdfd8f1a8ed58b \
- --hash=sha256:50480dcd74fa63b8e78235957d302d98d98d82ccbfac4c7e12108ba9ecbdba15 \
- --hash=sha256:518a48c2aab7ce596d3bf79d0e275661b846e86e4d0e7dec34712c30fe07f02a \
- --hash=sha256:559b38da23606e68681337ad74622c4dbba02254fc9cb4488a305dd5975c7eeb \
- --hash=sha256:578510d88c6229d735855e1f278aa305270438d36a05031dfaae5067cc8eb04d \
- --hash=sha256:597bd9c8419bc7c6af5604e55847789b69123bbe25d65cc6ad3012b4f3c98d8b \
- --hash=sha256:5a8eb7ed8d4198bccbd07058416eeec51686b498e784eda166395a23eb99138e \
- --hash=sha256:5c0dd1636633e7e6a0afe7bf6a51a14992b7f8e60de5789018ebbdfae55b040a \
- --hash=sha256:5cb1785d97b0c3d1d1a16bc1d710c4a0049daefc4935f3a8f31f827f4d3d2e7f \
- --hash=sha256:5d1f9575a12bed9e9eedd9a4972834b08c97a352bd17955ccdebfeca5913fa0a \
- --hash=sha256:5d8c41325b382c07799a3682c1c258469ea2ff97103c53717b7893862d0c98ce \
- --hash=sha256:5dae5f21afb91322f2ff791895ddd8889e5e947ff59f71b46041c8ce6db790bc \
- --hash=sha256:600fd103672b925fe62ed08e0d874ea34d692474df6f4bf7ebe148b30f89f39f \
- --hash=sha256:6408a7b064595afcab0a49393a413732a35788f2a5092fdc6266952ed67de586 \
- --hash=sha256:652a2c9ccfb556235b2b501a3a7cf3742148cd22e04b5625c5fe057ea3e3191f \
- --hash=sha256:665e1b916b043cef294bc54d47bf02d87e13f769bc4bc5fa225a24b3a6c5aca9 \
- --hash=sha256:691ab2ac363b8217f7d31b3497108fb1f50faab2f75dfb03284ec2f217e87bf8 \
- --hash=sha256:6c52f062424c523d6c4db85518774cc3d50f5539dd6eed32b8f6229b26f24d40 \
- --hash=sha256:6c6db3b84c87d48d0088943bf33440e0c42370b99b1c2a7989216f7b42eede60 \
- --hash=sha256:7311c0a0dcadb89b36b7025dfd8326ecfa36964e29913074d47382706e516a7c \
- --hash=sha256:7aac39bcf8d4770d089588a2e1dd111cbaa42df5a94be3114222057d68336bd0 \
- --hash=sha256:7b03048319bfc6170e93bd60728a1af51d3dd7704935feb228c4d4faab35d334 \
- --hash=sha256:7e7976bf1910a8116b523b9f9f58bf410f3e8aa330cd9a2bb2953f9266ab49af \
- --hash=sha256:8089c852a56c2966cf18835db62d9b34fef7ba74c726ad943928d494fa7f4735 \
- --hash=sha256:86172b0831b82ce4f7877f280055892b31179e1576aa00d0df3bb1bbf8c3e524 \
- --hash=sha256:89b54027a766529136a06cfebeecb3a04900397a3590fd252160b888479517bf \
- --hash=sha256:89c7e895002bbe49cdc5426150377cbbc04767d7547ed145473f496dfa40408b \
- --hash=sha256:8b7e5304e34942bf62e15184219a7b5ad4ff7f3bb5cca4d984f37df1a0e1aee2 \
- --hash=sha256:8fd420ef0c52c88b5a035a0886f367748c72147b2b8f384c9d12656678dfdfa9 \
- --hash=sha256:98edb152429ab62a1818039744d8fbb3ccab98a7c29fc3d5fcef158f3f1f68b7 \
- --hash=sha256:99c1506ea77c11531d75e3a412832a13a71c7ebc8192ab9e4b2e355555920e3e \
- --hash=sha256:9ad8fa5937ab05218e2b6a4cff30295ad35afd2f83ac592e68c0d871bb0fdbc4 \
- --hash=sha256:9f51079765661884a486727f0729d29054242f74b46186026582b4e4769918e4 \
- --hash=sha256:a003d7422449f6d1e3a34e3dd4110c22148336918ddbfc6a32581cd54b2e0b2b \
- --hash=sha256:a0b1cd6232e2b618adcc54d9882e4e662a089d5768cd188f7c245b4c8c44a397 \
- --hash=sha256:a285e3eb7a5a45a2ff504e31f4a8d1b12ef62e84e5411c6804a42197c1cf586c \
- --hash=sha256:a37691702ed687799de29a518d63d4682d9016932db66d4e90c345831b02fb4e \
- --hash=sha256:a550ae29b95c6dc13cf69e2c9dc5747f814c54eeb2e32d683e5e93af56caa029 \
- --hash=sha256:ab174cd7d29a62dd139c44bf74b698039328f45cb03b4596c43473a46656b2f3 \
- --hash=sha256:ab323b787d6e18b3d91a72fc99b1a2c28651e4358749842b8f8dfacd28ef2052 \
- --hash=sha256:adebb5bee0f0af4909c30db0d890c773d1a92ffe83da908e2e9e720f8edf3984 \
- --hash=sha256:aee2810642b2898bb187ced9b349e95d2a7272930796e022efaf12e99dccd293 \
- --hash=sha256:af9a332e572978f0218686636610555ae3defd1633597be015ed50289a03c523 \
- --hash=sha256:b574c51cf7d5d62e9be37ba446224b59a2da26dc4c1bb2ecbe936a4fb1a7cb7f \
- --hash=sha256:b66e95d05ba806247aaa1561f080abc7975daf715c30780ff92a20e4ec546e1b \
- --hash=sha256:b81b5e3511211631b3f672a595e3221252c90af017e399056d0faabb9538aa80 \
- --hash=sha256:b957b71c6b2387610f556a7eb0828afbe40b4a98036fc0d2acfa5a44a0c2036f \
- --hash=sha256:bb66b7cc26f50977108790e2456b7921e773f23db5630261102233eb355a3b79 \
- --hash=sha256:c6008de247150668a705a6338156efb92334113421ceecf7438a12c9a12dab23 \
- --hash=sha256:c7697918b5be27424e9ce568193efd13d925c4481dd364e43f5dff72d33e10f8 \
- --hash=sha256:cb9bb857b2d057c6dfc72ac5f3b44836924ba15721882ef103cecb40d002d80e \
- --hash=sha256:cc7d296b5ea4d29e6570dabeaed58d31c3fea35a633a69679fb03d7664f43fb3 \
- --hash=sha256:d242e8ac078781f1de88bf823d70c1a9b3c7950a44cdf4b7c012e22ccbcd8e4e \
- --hash=sha256:d2912fd8114fc5545aa3a4b5576512f64c55a03f3ebcca4c10194d593d43ea36 \
- --hash=sha256:d470ab1178551dd17fdba0fef463359c41aaa613cdcd7ff8373f54be629f9f8f \
- --hash=sha256:d4ce8e329c93845720cd2014659ca67eac35f6433fd3050393d85f3ecef0dad5 \
- --hash=sha256:d6e4571eedf43af33d0fc233a382a76e849badbccdf1ac438841308652a08e1f \
- --hash=sha256:e65498daf4b583091ccbb2556c7000abf0f3349fcd57ef7adc9a84a394ed29f6 \
- --hash=sha256:e879bb6cd5c73848ef3b2b48b8af9ff08c5b71ecda8048b7dd22d8a33f60be32 \
- --hash=sha256:e9e8064fb1cc019296958595f6db671fba95209e3ceb0c4734c9baf97de04b20 \
- --hash=sha256:f7ed2c6543bad5a7d5530eb9e78c53132f93dfa44a28492db88b41cdab885202 \
- --hash=sha256:f95c00d5d6700b2b890479664a06e754974848afaae5e21beb4d83c106923fd0 \
- --hash=sha256:f975aa7ef9684ce7e2c18a3aa8f8e2106ce1e46b94ab713d156b2898811651d3 \
- --hash=sha256:fbfa2a7c10cc2623f412753cddf391c7f971c52ca40a3f65dc5039b2939e8563 \
- --hash=sha256:fc354a04072b765eccf2204f588a7a532c9511e8b9c7f900e1b64e3e33487090 \
- --hash=sha256:fc44ef1f3de4f45b50ccf9136999d71abb99dca7706bc75d222ed350b9fd2289
-pip==26.0.1 \
- --hash=sha256:bdb1b08f4274833d62c1aa29e20907365a2ceb950410df15fc9521bad440122b \
- --hash=sha256:c4037d8a277c89b320abe636d59f91e6d0922d08a05b60e85e53b296613346d8
-platformdirs==4.5.1 \
- --hash=sha256:61d5cdcc6065745cdd94f0f878977f8de9437be93de97c1c12f853c9c0cdcbda \
- --hash=sha256:d03afa3963c806a9bed9d5125c8f4cb2fdaf74a55ab60e5d59b3fde758104d31
-protobuf==5.29.6 \
- --hash=sha256:36ade6ff88212e91aef4e687a971a11d7d24d6948a66751abc1b3238648f5d05 \
- --hash=sha256:62e8a3114992c7c647bce37dcc93647575fc52d50e48de30c6fcb28a6a291eb1 \
- --hash=sha256:6b9edb641441b2da9fa8f428760fc136a49cf97a52076010cf22a2ff73438a86 \
- --hash=sha256:76e07e6567f8baf827137e8d5b8204b6c7b6488bbbff1bf0a72b383f77999c18 \
- --hash=sha256:7e6ad413275be172f67fdee0f43484b6de5a904cc1c3ea9804cb6fe2ff366eda \
- --hash=sha256:831e2da16b6cc9d8f1654c041dd594eda43391affd3c03a91bea7f7f6da106d6 \
- --hash=sha256:a8866b2cff111f0f863c1b3b9e7572dc7eaea23a7fae27f6fc613304046483e6 \
- --hash=sha256:b5a169e664b4057183a34bdc424540e86eea47560f3c123a0d64de4e137f9269 \
- --hash=sha256:cb4c86de9cd8a7f3a256b9744220d87b847371c6b2f10bde87768918ef33ba49 \
- --hash=sha256:da9ee6a5424b6b30fd5e45c5ea663aef540ca95f9ad99d1e887e819cdf9b8723 \
- --hash=sha256:e3387f44798ac1106af0233c04fb8abf543772ff241169946f698b3a9a3d3ab9
-psutil==7.2.1 \
- --hash=sha256:05cc68dbb8c174828624062e73078e7e35406f4ca2d0866c272c2410d8ef06d1 \
- --hash=sha256:08a2f175e48a898c8eb8eace45ce01777f4785bc744c90aa2cc7f2fa5462a266 \
- --hash=sha256:0d67c1822c355aa6f7314d92018fb4268a76668a536f133599b91edd48759442 \
- --hash=sha256:2ceae842a78d1603753561132d5ad1b2f8a7979cb0c283f5b52fb4e6e14b1a79 \
- --hash=sha256:35630d5af80d5d0d49cfc4d64c1c13838baf6717a13effb35869a5919b854cdf \
- --hash=sha256:3fce5f92c22b00cdefd1645aa58ab4877a01679e901555067b1bd77039aa589f \
- --hash=sha256:494c513ccc53225ae23eec7fe6e1482f1b8a44674241b54561f755a898650679 \
- --hash=sha256:5e38404ca2bb30ed7267a46c02f06ff842e92da3bb8c5bfdadbd35a5722314d8 \
- --hash=sha256:81442dac7abfc2f4f4385ea9e12ddf5a796721c0f6133260687fec5c3780fa49 \
- --hash=sha256:923f8653416604e356073e6e0bccbe7c09990acef442def2f5640dd0faa9689f \
- --hash=sha256:93f3f7b0bb07711b49626e7940d6fe52aa9940ad86e8f7e74842e73189712129 \
- --hash=sha256:99a4cd17a5fdd1f3d014396502daa70b5ec21bf4ffe38393e152f8e449757d67 \
- --hash=sha256:ab2b98c9fc19f13f59628d94df5cc4cc4844bc572467d113a8b517d634e362c6 \
- --hash=sha256:b1b0671619343aa71c20ff9767eced0483e4fc9e1f489d50923738caf6a03c17 \
- --hash=sha256:b2e953fcfaedcfbc952b44744f22d16575d3aa78eb4f51ae74165b4e96e55f42 \
- --hash=sha256:ba9f33bb525b14c3ea563b2fd521a84d2fa214ec59e3e6a2858f78d0844dd60d \
- --hash=sha256:cfbe6b40ca48019a51827f20d830887b3107a74a79b01ceb8cc8de4ccb17b672 \
- --hash=sha256:d34d2ca888208eea2b5c68186841336a7f5e0b990edec929be909353a202768a \
- --hash=sha256:ea46c0d060491051d39f0d2cff4f98d5c72b288289f57a21556cc7d504db37fc \
- --hash=sha256:f7583aec590485b43ca601dd9cea0dcd65bd7bb21d30ef4ddbf4ea6b5ed1bdd3 \
- --hash=sha256:f78baafb38436d5a128f837fab2d92c276dfb48af01a240b861ae02b2413ada8
-pycparser==3.0 \
- --hash=sha256:600f49d217304a5902ac3c37e1281c9fe94e4d0489de643a9504c5cdfdfc6b29 \
- --hash=sha256:b727414169a36b7d524c1c3e31839a521725078d7b2ff038656844266160a992
-pydot==3.0.4 \
- --hash=sha256:3ce88b2558f3808b0376f22bfa6c263909e1c3981e2a7b629b65b451eee4a25d \
- --hash=sha256:bfa9c3fc0c44ba1d132adce131802d7df00429d1a79cc0346b0a5cd374dbe9c6
-pygments==2.20.0 \
- --hash=sha256:6757cd03768053ff99f3039c1a36d6c0aa0b263438fcab17520b30a303a82b5f \
- --hash=sha256:81a9e26dd42fd28a23a2d169d86d7ac03b46e2f8b59ed4698fb4785f946d0176
-pymoo==0.6.1.6 \
- --hash=sha256:007bdabe638d5cc1ff98d1df096e7cea42f266ecc7b84fc0aac695a20b6f02f6 \
- --hash=sha256:01e848762e04d183e8dcc12e1cfb1e2cdd2c5efeda47961dbcf6c86c09751e30 \
- --hash=sha256:2a2e9524b07baef135a7143700cc65a504c2123c717a6924fa939947b4fdba73 \
- --hash=sha256:38fa5bf507cd7c07eae0286948db79f761b5b4c5d5956a31fd0e5a771c83fa47 \
- --hash=sha256:3b9dcb6959cf2b12c0f1d4ac971fa53074430eb976367b2e7eec0d0301423f31 \
- --hash=sha256:466f88073c9f498bcf75334a2e47279968003d91b7fe01df1b056b5a6c3434d8 \
- --hash=sha256:49a9cde6ff7cc8497458166e2c14734bf4879c76256412cce0de2af86d2b71a8 \
- --hash=sha256:50472dfb93b90b4801c63070a4afc786c52f7f56b345f345024b39cd35ff06c4 \
- --hash=sha256:5086f46020aa1a13a0e3142ef87b6487e91ea5d25b7b69467039f0d3f5a57cb2 \
- --hash=sha256:533de86c284bf3b7b6da871e2ae07a914eaa5d3a51f5c4e6d7c055d3471c4467 \
- --hash=sha256:5607e0359c96691158192bd3e2b939beab6d4ed25032ef4cb79edf588c2bb475 \
- --hash=sha256:56ecb6f9f5ac0559829e183a23ea3672eae8c1eb9d745305bef98da8d1614d28 \
- --hash=sha256:62599637724f74820e16d813e6ca86af1733add4251e22b5a6c677df75dec06b \
- --hash=sha256:642308eae98c5b8aee96272e10c6409dd97289ef214ec7383b39425f01549b5f \
- --hash=sha256:718b9e971e436c2ded520c4a6fe8c83b7f305da1fcc539d6386d025c6e9bfe66 \
- --hash=sha256:791188677b42e104c41cd1c65370a94fbbec756d8894947863f42a1f76352c90 \
- --hash=sha256:798e5af41527324303e56ac806f9ec34c65f31c2df9e904daeca68027bf91266 \
- --hash=sha256:7c042120df86240c1451e2a1afb0f76b64acb8d736fb7687bc5e4aec40f7ff9e \
- --hash=sha256:7d13e14434f0555fbeda20764d34be4dc94f53ae7040a337928d7bfe42d3422a \
- --hash=sha256:819a27ff19c146a64853a86ebab02bc6bbf28fb9fcf119e93cfc855ec6314f72 \
- --hash=sha256:981bd5b8581e959814a0e51982be1a964945fb7f599cfe7857a63cbc1ad97d58 \
- --hash=sha256:a4f590fc62021d2583afdc88a2eb5ec878699786585f9ca82edf1fc32557ac96 \
- --hash=sha256:a68fb9d807c1cb6f36e13c24731c496b1e98b9c8810125072f19bd56baa6a4d5 \
- --hash=sha256:b3e990dea0ad84fa2f89462215e34960579dbb9fef4d880aecb09f50cc18e713 \
- --hash=sha256:b9dd7da8b18bffac6b09c75b4a4a645232994d91fc0cdfe39baffdc08fa5d252 \
- --hash=sha256:c16e05b36e081d0ea6aee499346f65466811f63f5837f0d09de866624cb7355b \
- --hash=sha256:c8cfd085b74b3f673378109d8d0734d0cab460e3e78c09f691120ac0c2698aad \
- --hash=sha256:cd2b0fb19ce3e71a177ed3b965ce2d4334ffb6aad8f621c124de6dafa09b7758 \
- --hash=sha256:d48077c7b612b149e7db5351459bf96a0950e84ebcd5b7b953bf46b3dcf1ac28 \
- --hash=sha256:df9520fd7fab0761d5698dd36eabcfb18b5eb970ddf01e10eb30b764e4bb72af \
- --hash=sha256:e0a42b6f3c26b2725b1edb1d4cfb98e5640bb0638813080ba1aa0daf23b2fd05 \
- --hash=sha256:ecf1ce515e72d59b78c909aa142d2369cc3c6d771ac7c9162831c0b62a869c8f \
- --hash=sha256:f2d48fa7f8f8ea314405ad49a6b7f3dcf8f12c2024e319e2d47ede0937f8b39b \
- --hash=sha256:f41b359d6de6c9d99ab7da94f115c5aea4d21e74a252bef7c8debec8f20f0b9f \
- --hash=sha256:f62fb1e2ba15f22ff5569c6d364dbf9dfe8b3722a114e6bc1df3cc10f8dd5044 \
- --hash=sha256:fd4715b4015b10ad3cf02e5eb9abb57fe4bfa4d3378f9076b76d60c1ddfa46d6
-pyparsing==3.3.2 \
- --hash=sha256:850ba148bd908d7e2411587e247a1e4f0327839c40e2e5e6d05a007ecc69911d \
- --hash=sha256:c777f4d763f140633dcb6d8a3eda953bf7a214dc4eff598413c070bcdc117cbc
-python-dateutil==2.9.0.post0 \
- --hash=sha256:37dd54208da7e1cd875388217d5e00ebd4179249f90fb72437e91a35459a0ad3 \
- --hash=sha256:a8b2bc7bffae282281c8140a97d3aa9c14da0b136dfe83f850eea9a5f7470427
-pytz==2025.2 \
- --hash=sha256:360b9e3dbb49a209c21ad61809c7fb453643e048b38924c765813546746e81c3 \
- --hash=sha256:5ddf76296dd8c44c26eb8f4b6f35488f3ccbf6fbbd7adee0b7262d43f0ec2f00
-referencing==0.37.0 \
- --hash=sha256:381329a9f99628c9069361716891d34ad94af76e461dcb0335825aecc7692231 \
- --hash=sha256:44aefc3142c5b842538163acb373e24cce6632bd54bdb01b21ad5863489f50d8
-rich==14.2.0 \
- --hash=sha256:73ff50c7c0c1c77c8243079283f4edb376f0f6442433aecb8ce7e6d0b92d1fe4 \
- --hash=sha256:76bc51fe2e57d2b1be1f96c524b890b816e334ab4c1e45888799bfaab0021edd
-rpds-py==0.30.0 \
- --hash=sha256:07ae8a593e1c3c6b82ca3292efbe73c30b61332fd612e05abee07c79359f292f \
- --hash=sha256:0a59119fc6e3f460315fe9d08149f8102aa322299deaa5cab5b40092345c2136 \
- --hash=sha256:0c0e95f6819a19965ff420f65578bacb0b00f251fefe2c8b23347c37174271f3 \
- --hash=sha256:0d08f00679177226c4cb8c5265012eea897c8ca3b93f429e546600c971bcbae7 \
- --hash=sha256:0ed177ed9bded28f8deb6ab40c183cd1192aa0de40c12f38be4d59cd33cb5c65 \
- --hash=sha256:12f90dd7557b6bd57f40abe7747e81e0c0b119bef015ea7726e69fe550e394a4 \
- --hash=sha256:1726859cd0de969f88dc8673bdd954185b9104e05806be64bcd87badbe313169 \
- --hash=sha256:1ab5b83dbcf55acc8b08fc62b796ef672c457b17dbd7820a11d6c52c06839bdf \
- --hash=sha256:1b151685b23929ab7beec71080a8889d4d6d9fa9a983d213f07121205d48e2c4 \
- --hash=sha256:1f3587eb9b17f3789ad50824084fa6f81921bbf9a795826570bda82cb3ed91f2 \
- --hash=sha256:250fa00e9543ac9b97ac258bd37367ff5256666122c2d0f2bc97577c60a1818c \
- --hash=sha256:2771c6c15973347f50fece41fc447c054b7ac2ae0502388ce3b6738cd366e3d4 \
- --hash=sha256:27f4b0e92de5bfbc6f86e43959e6edd1425c33b5e69aab0984a72047f2bcf1e3 \
- --hash=sha256:2e6ecb5a5bcacf59c3f912155044479af1d0b6681280048b338b28e364aca1f6 \
- --hash=sha256:32c8528634e1bf7121f3de08fa85b138f4e0dc47657866630611b03967f041d7 \
- --hash=sha256:33f559f3104504506a44bb666b93a33f5d33133765b0c216a5bf2f1e1503af89 \
- --hash=sha256:3896fa1be39912cf0757753826bc8bdc8ca331a28a7c4ae46b7a21280b06bb85 \
- --hash=sha256:389a2d49eded1896c3d48b0136ead37c48e221b391c052fba3f4055c367f60a6 \
- --hash=sha256:39c02563fc592411c2c61d26b6c5fe1e51eaa44a75aa2c8735ca88b0d9599daa \
- --hash=sha256:3adbb8179ce342d235c31ab8ec511e66c73faa27a47e076ccc92421add53e2bb \
- --hash=sha256:3d4a69de7a3e50ffc214ae16d79d8fbb0922972da0356dcf4d0fdca2878559c6 \
- --hash=sha256:3e62880792319dbeb7eb866547f2e35973289e7d5696c6e295476448f5b63c87 \
- --hash=sha256:3e8eeb0544f2eb0d2581774be4c3410356eba189529a6b3e36bbbf9696175856 \
- --hash=sha256:422c3cb9856d80b09d30d2eb255d0754b23e090034e1deb4083f8004bd0761e4 \
- --hash=sha256:4559c972db3a360808309e06a74628b95eaccbf961c335c8fe0d590cf587456f \
- --hash=sha256:46e83c697b1f1c72b50e5ee5adb4353eef7406fb3f2043d64c33f20ad1c2fc53 \
- --hash=sha256:47b0ef6231c58f506ef0b74d44e330405caa8428e770fec25329ed2cb971a229 \
- --hash=sha256:47e77dc9822d3ad616c3d5759ea5631a75e5809d5a28707744ef79d7a1bcfcad \
- --hash=sha256:47f236970bccb2233267d89173d3ad2703cd36a0e2a6e92d0560d333871a3d23 \
- --hash=sha256:47f9a91efc418b54fb8190a6b4aa7813a23fb79c51f4bb84e418f5476c38b8db \
- --hash=sha256:495aeca4b93d465efde585977365187149e75383ad2684f81519f504f5c13038 \
- --hash=sha256:4c5f36a861bc4b7da6516dbdf302c55313afa09b81931e8280361a4f6c9a2d27 \
- --hash=sha256:4cc2206b76b4f576934f0ed374b10d7ca5f457858b157ca52064bdfc26b9fc00 \
- --hash=sha256:4e7fc54e0900ab35d041b0601431b0a0eb495f0851a0639b6ef90f7741b39a18 \
- --hash=sha256:51a1234d8febafdfd33a42d97da7a43f5dcb120c1060e352a3fbc0c6d36e2083 \
- --hash=sha256:55f66022632205940f1827effeff17c4fa7ae1953d2b74a8581baaefb7d16f8c \
- --hash=sha256:58edca431fb9b29950807e301826586e5bbf24163677732429770a697ffe6738 \
- --hash=sha256:5965af57d5848192c13534f90f9dd16464f3c37aaf166cc1da1cae1fd5a34898 \
- --hash=sha256:5ba103fb455be00f3b1c2076c9d4264bfcb037c976167a6047ed82f23153f02e \
- --hash=sha256:5d4c2aa7c50ad4728a094ebd5eb46c452e9cb7edbfdb18f9e1221f597a73e1e7 \
- --hash=sha256:61046904275472a76c8c90c9ccee9013d70a6d0f73eecefd38c1ae7c39045a08 \
- --hash=sha256:613aa4771c99f03346e54c3f038e4cc574ac09a3ddfb0e8878487335e96dead6 \
- --hash=sha256:626a7433c34566535b6e56a1b39a7b17ba961e97ce3b80ec62e6f1312c025551 \
- --hash=sha256:669b1805bd639dd2989b281be2cfd951c6121b65e729d9b843e9639ef1fd555e \
- --hash=sha256:679ae98e00c0e8d68a7fda324e16b90fd5260945b45d3b824c892cec9eea3288 \
- --hash=sha256:67b02ec25ba7a9e8fa74c63b6ca44cf5707f2fbfadae3ee8e7494297d56aa9df \
- --hash=sha256:68f19c879420aa08f61203801423f6cd5ac5f0ac4ac82a2368a9fcd6a9a075e0 \
- --hash=sha256:692bef75a5525db97318e8cd061542b5a79812d711ea03dbc1f6f8dbb0c5f0d2 \
- --hash=sha256:6abc8880d9d036ecaafe709079969f56e876fcf107f7a8e9920ba6d5a3878d05 \
- --hash=sha256:6bdfdb946967d816e6adf9a3d8201bfad269c67efe6cefd7093ef959683c8de0 \
- --hash=sha256:6de2a32a1665b93233cde140ff8b3467bdb9e2af2b91079f0333a0974d12d464 \
- --hash=sha256:73c67f2db7bc334e518d097c6d1e6fed021bbc9b7d678d6cc433478365d1d5f5 \
- --hash=sha256:74a3243a411126362712ee1524dfc90c650a503502f135d54d1b352bd01f2404 \
- --hash=sha256:76fec018282b4ead0364022e3c54b60bf368b9d926877957a8624b58419169b7 \
- --hash=sha256:7c64d38fb49b6cdeda16ab49e35fe0da2e1e9b34bc38bd78386530f218b37139 \
- --hash=sha256:7cee9c752c0364588353e627da8a7e808a66873672bcb5f52890c33fd965b394 \
- --hash=sha256:7e6ecfcb62edfd632e56983964e6884851786443739dbfe3582947e87274f7cb \
- --hash=sha256:806f36b1b605e2d6a72716f321f20036b9489d29c51c91f4dd29a3e3afb73b15 \
- --hash=sha256:858738e9c32147f78b3ac24dc0edb6610000e56dc0f700fd5f651d0a0f0eb9ff \
- --hash=sha256:8d6d1cc13664ec13c1b84241204ff3b12f9bb82464b8ad6e7a5d3486975c2eed \
- --hash=sha256:9027da1ce107104c50c81383cae773ef5c24d296dd11c99e2629dbd7967a20c6 \
- --hash=sha256:922e10f31f303c7c920da8981051ff6d8c1a56207dbdf330d9047f6d30b70e5e \
- --hash=sha256:945dccface01af02675628334f7cf49c2af4c1c904748efc5cf7bbdf0b579f95 \
- --hash=sha256:946fe926af6e44f3697abbc305ea168c2c31d3e3ef1058cf68f379bf0335a78d \
- --hash=sha256:95f0802447ac2d10bcc69f6dc28fe95fdf17940367b21d34e34c737870758950 \
- --hash=sha256:9854cf4f488b3d57b9aaeb105f06d78e5529d3145b1e4a41750167e8c213c6d3 \
- --hash=sha256:993914b8e560023bc0a8bf742c5f303551992dcb85e247b1e5c7f4a7d145bda5 \
- --hash=sha256:99b47d6ad9a6da00bec6aabe5a6279ecd3c06a329d4aa4771034a21e335c3a97 \
- --hash=sha256:9a4e86e34e9ab6b667c27f3211ca48f73dba7cd3d90f8d5b11be56e5dbc3fb4e \
- --hash=sha256:9cf69cdda1f5968a30a359aba2f7f9aa648a9ce4b580d6826437f2b291cfc86e \
- --hash=sha256:a090322ca841abd453d43456ac34db46e8b05fd9b3b4ac0c78bcde8b089f959b \
- --hash=sha256:a1010ed9524c73b94d15919ca4d41d8780980e1765babf85f9a2f90d247153dd \
- --hash=sha256:a161f20d9a43006833cd7068375a94d035714d73a172b681d8881820600abfad \
- --hash=sha256:a1d0bc22a7cdc173fedebb73ef81e07faef93692b8c1ad3733b67e31e1b6e1b8 \
- --hash=sha256:a2bffea6a4ca9f01b3f8e548302470306689684e61602aa3d141e34da06cf425 \
- --hash=sha256:a452763cc5198f2f98898eb98f7569649fe5da666c2dc6b5ddb10fde5a574221 \
- --hash=sha256:a4796a717bf12b9da9d3ad002519a86063dcac8988b030e405704ef7d74d2d9d \
- --hash=sha256:a51033ff701fca756439d641c0ad09a41d9242fa69121c7d8769604a0a629825 \
- --hash=sha256:a8fa71a2e078c527c3e9dc9fc5a98c9db40bcc8a92b4e8858e36d329f8684b51 \
- --hash=sha256:ac37f9f516c51e5753f27dfdef11a88330f04de2d564be3991384b2f3535d02e \
- --hash=sha256:ac98b175585ecf4c0348fd7b29c3864bda53b805c773cbf7bfdaffc8070c976f \
- --hash=sha256:acd7eb3f4471577b9b5a41baf02a978e8bdeb08b4b355273994f8b87032000a8 \
- --hash=sha256:ad1fa8db769b76ea911cb4e10f049d80bf518c104f15b3edb2371cc65375c46f \
- --hash=sha256:b40fb160a2db369a194cb27943582b38f79fc4887291417685f3ad693c5a1d5d \
- --hash=sha256:b4dc1a6ff022ff85ecafef7979a2c6eb423430e05f1165d6688234e62ba99a07 \
- --hash=sha256:ba3af48635eb83d03f6c9735dfb21785303e73d22ad03d489e88adae6eab8877 \
- --hash=sha256:ba81a9203d07805435eb06f536d95a266c21e5b2dfbf6517748ca40c98d19e31 \
- --hash=sha256:c2262bdba0ad4fc6fb5545660673925c2d2a5d9e2e0fb603aad545427be0fc58 \
- --hash=sha256:c77afbd5f5250bf27bf516c7c4a016813eb2d3e116139aed0096940c5982da94 \
- --hash=sha256:ca28829ae5f5d569bb62a79512c842a03a12576375d5ece7d2cadf8abe96ec28 \
- --hash=sha256:cdc62c8286ba9bf7f47befdcea13ea0e26bf294bda99758fd90535cbaf408000 \
- --hash=sha256:d948b135c4693daff7bc2dcfc4ec57237a29bd37e60c2fabf5aff2bbacf3e2f1 \
- --hash=sha256:d96c2086587c7c30d44f31f42eae4eac89b60dabbac18c7669be3700f13c3ce1 \
- --hash=sha256:d9a0ca5da0386dee0655b4ccdf46119df60e0f10da268d04fe7cc87886872ba7 \
- --hash=sha256:da279aa314f00acbb803da1e76fa18666778e8a8f83484fba94526da5de2cba7 \
- --hash=sha256:dbd936cde57abfee19ab3213cf9c26be06d60750e60a8e4dd85d1ab12c8b1f40 \
- --hash=sha256:dc4f992dfe1e2bc3ebc7444f6c7051b4bc13cd8e33e43511e8ffd13bf407010d \
- --hash=sha256:dc824125c72246d924f7f796b4f63c1e9dc810c7d9e2355864b3c3a73d59ade0 \
- --hash=sha256:dd8ff7cf90014af0c0f787eea34794ebf6415242ee1d6fa91eaba725cc441e84 \
- --hash=sha256:dea5b552272a944763b34394d04577cf0f9bd013207bc32323b5a89a53cf9c2f \
- --hash=sha256:dff13836529b921e22f15cb099751209a60009731a68519630a24d61f0b1b30a \
- --hash=sha256:e0b65193a413ccc930671c55153a03ee57cecb49e6227204b04fae512eb657a7 \
- --hash=sha256:e5d3e6b26f2c785d65cc25ef1e5267ccbe1b069c5c21b8cc724efee290554419 \
- --hash=sha256:e7536cd91353c5273434b4e003cbda89034d67e7710eab8761fd918ec6c69cf8 \
- --hash=sha256:eb0b93f2e5c2189ee831ee43f156ed34e2a89a78a66b98cadad955972548be5a \
- --hash=sha256:eb2c4071ab598733724c08221091e8d80e89064cd472819285a9ab0f24bcedb9 \
- --hash=sha256:ec7c4490c672c1a0389d319b3a9cfcd098dcdc4783991553c332a15acf7249be \
- --hash=sha256:ee454b2a007d57363c2dfd5b6ca4a5d7e2c518938f8ed3b706e37e5d470801ed \
- --hash=sha256:ee6af14263f25eedc3bb918a3c04245106a42dfd4f5c2285ea6f997b1fc3f89a \
- --hash=sha256:f14fc5df50a716f7ece6a80b6c78bb35ea2ca47c499e422aa4463455dd96d56d \
- --hash=sha256:f207f69853edd6f6700b86efb84999651baf3789e78a466431df1331608e5324 \
- --hash=sha256:f251c812357a3fed308d684a5079ddfb9d933860fc6de89f2b7ab00da481e65f \
- --hash=sha256:f83424d738204d9770830d35290ff3273fbb02b41f919870479fab14b9d303b2 \
- --hash=sha256:f8d1736cfb49381ba528cd5baa46f82fdc65c06e843dab24dd70b63d09121b3f \
- --hash=sha256:fe5fa731a1fa8a0a56b0977413f8cacac1768dad38d16b3a296712709476fbd5
-safetensors==0.7.0 \
- --hash=sha256:0071bffba4150c2f46cae1432d31995d77acfd9f8db598b5d1a2ce67e8440ad2 \
- --hash=sha256:07663963b67e8bd9f0b8ad15bb9163606cd27cc5a1b96235a50d8369803b96b0 \
- --hash=sha256:12f49080303fa6bb424b362149a12949dfbbf1e06811a88f2307276b0c131afd \
- --hash=sha256:1d060c70284127fa805085d8f10fbd0962792aed71879d00864acda69dbab981 \
- --hash=sha256:42cb091236206bb2016d245c377ed383aa7f78691748f3bb6ee1bfa51ae2ce6a \
- --hash=sha256:473b32699f4200e69801bf5abf93f1a4ecd432a70984df164fc22ccf39c4a6f3 \
- --hash=sha256:54bef08bf00a2bff599982f6b08e8770e09cc012d7bba00783fc7ea38f1fb37d \
- --hash=sha256:5d72abdb8a4d56d4020713724ba81dac065fedb7f3667151c4a637f1d3fb26c0 \
- --hash=sha256:672132907fcad9f2aedcb705b2d7b3b93354a2aec1b2f706c4db852abe338f85 \
- --hash=sha256:6999421eb8ba9df4450a16d9184fcb7bef26240b9f98e95401f17af6c2210b71 \
- --hash=sha256:7b95a3fa7b3abb9b5b0e07668e808364d0d40f6bbbf9ae0faa8b5b210c97b140 \
- --hash=sha256:8469155f4cb518bafb4acf4865e8bb9d6804110d2d9bdcaa78564b9fd841e104 \
- --hash=sha256:94fd4858284736bb67a897a41608b5b0c2496c9bdb3bf2af1fa3409127f20d57 \
- --hash=sha256:b0f6d66c1c538d5a94a73aa9ddca8ccc4227e6c9ff555322ea40bdd142391dd4 \
- --hash=sha256:c74af94bf3ac15ac4d0f2a7c7b4663a15f8c2ab15ed0fc7531ca61d0835eccba \
- --hash=sha256:c82f4d474cf725255d9e6acf17252991c3c8aac038d6ef363a4bf8be2f6db517 \
- --hash=sha256:cdab83a366799fa730f90a4ebb563e494f28e9e92c4819e556152ad55e43591b \
- --hash=sha256:cfdead2f57330d76aa7234051dadfa7d4eedc0e5a27fd08e6f96714a92b00f09 \
- --hash=sha256:d1239932053f56f3456f32eb9625590cc7582e905021f94636202a864d470755 \
- --hash=sha256:dac7252938f0696ddea46f5e855dd3138444e82236e3be475f54929f0c510d48 \
- --hash=sha256:dc92bc2db7b45bda4510e4f51c59b00fe80b2d6be88928346e4294ce1c2abe7c \
- --hash=sha256:e07d91d0c92a31200f25351f4acb2bc6aff7f48094e13ebb1d0fb995b54b6542 \
- --hash=sha256:f4729811a6640d019a4b7ba8638ee2fd21fa5ca8c7e7bdf0fed62068fcaac737
-scikit-learn==1.7.2 \
- --hash=sha256:0486c8f827c2e7b64837c731c8feff72c0bd2b998067a8a9cbc10643c31f0fe1 \
- --hash=sha256:0b7dacaa05e5d76759fb071558a8b5130f4845166d88654a0f9bdf3eb57851b7 \
- --hash=sha256:191e5550980d45449126e23ed1d5e9e24b2c68329ee1f691a3987476e115e09c \
- --hash=sha256:20e9e49ecd130598f1ca38a1d85090e1a600147b9c02fa6f15d69cb53d968fda \
- --hash=sha256:2a41e2a0ef45063e654152ec9d8bcfc39f7afce35b08902bfe290c2498a67a6a \
- --hash=sha256:36749fb62b3d961b1ce4fedf08fa57a1986cd409eff2d783bca5d4b9b5fce51c \
- --hash=sha256:4a847fea807e278f821a0406ca01e387f97653e284ecbd9750e3ee7c90347f18 \
- --hash=sha256:502c18e39849c0ea1a5d681af1dbcf15f6cce601aebb657aabbfe84133c1907f \
- --hash=sha256:57dc4deb1d3762c75d685507fbd0bc17160144b2f2ba4ccea5dc285ab0d0e973 \
- --hash=sha256:6088aa475f0785e01bcf8529f55280a3d7d298679f50c0bb70a2364a82d0b290 \
- --hash=sha256:63a9afd6f7b229aad94618c01c252ce9e6fa97918c5ca19c9a17a087d819440c \
- --hash=sha256:6b33579c10a3081d076ab403df4a4190da4f4432d443521674637677dc91e61f \
- --hash=sha256:7a4c328a71785382fe3fe676a9ecf2c86189249beff90bf85e22bdb7efaf9ae0 \
- --hash=sha256:7a58814265dfc52b3295b1900cfb5701589d30a8bb026c7540f1e9d3499d5ec8 \
- --hash=sha256:89877e19a80c7b11a2891a27c21c4894fb18e2c2e077815bcade10d34287b20d \
- --hash=sha256:8d91a97fa2b706943822398ab943cde71858a50245e31bc71dba62aab1d60a96 \
- --hash=sha256:8da8bf89d4d79aaec192d2bda62f9b56ae4e5b4ef93b6a56b5de4977e375c1f1 \
- --hash=sha256:9656e4a53e54578ad10a434dc1f993330568cfee176dff07112b8785fb413106 \
- --hash=sha256:96dc05a854add0e50d3f47a1ef21a10a595016da5b007c7d9cd9d0bffd1fcc61 \
- --hash=sha256:98335fb98509b73385b3ab2bd0639b1f610541d3988ee675c670371d6a87aa7c \
- --hash=sha256:9acb6c5e867447b4e1390930e3944a005e2cb115922e693c08a323421a6966e8 \
- --hash=sha256:9b7ed8d58725030568523e937c43e56bc01cadb478fc43c042a9aca1dacb3ba1 \
- --hash=sha256:abebbd61ad9e1deed54cca45caea8ad5f79e1b93173dece40bb8e0c658dbe6fe \
- --hash=sha256:acbc0f5fd2edd3432a22c69bed78e837c70cf896cd7993d71d51ba6708507476 \
- --hash=sha256:b4d6e9deed1a47aca9fe2f267ab8e8fe82ee20b4526b2c0cd9e135cea10feb44 \
- --hash=sha256:bb24510ed3f9f61476181e4db51ce801e2ba37541def12dc9333b946fc7a9cf8 \
- --hash=sha256:c7509693451651cd7361d30ce4e86a1347493554f172b1c72a39300fa2aea79e \
- --hash=sha256:ca250e6836d10e6f402436d6463d6c0e4d8e0234cfb6a9a47835bd392b852ce5 \
- --hash=sha256:e5bf3d930aee75a65478df91ac1225ff89cd28e9ac7bd1196853a9229b6adb0b \
- --hash=sha256:f95dc55b7902b91331fa4e5845dd5bde0580c9cd9612b1b2791b7e80c3d32615 \
- --hash=sha256:fa8f63940e29c82d1e67a45d5297bdebbcb585f5a5a50c4914cc2e852ab77f33
-scipy==1.15.3 \
- --hash=sha256:05dc6abcd105e1a29f95eada46d4a3f251743cfd7d3ae8ddb4088047f24ea477 \
- --hash=sha256:06efcba926324df1696931a57a176c80848ccd67ce6ad020c810736bfd58eb1c \
- --hash=sha256:0a769105537aa07a69468a0eefcd121be52006db61cdd8cac8a0e68980bbb723 \
- --hash=sha256:0bdd905264c0c9cfa74a4772cdb2070171790381a5c4d312c973382fc6eaf730 \
- --hash=sha256:0ff17c0bb1cb32952c09217d8d1eed9b53d1463e5f1dd6052c7857f83127d539 \
- --hash=sha256:14ed70039d182f411ffc74789a16df3835e05dc469b898233a245cdfd7f162cb \
- --hash=sha256:185cd3d6d05ca4b44a8f1595af87f9c372bb6acf9c808e99aa3e9aa03bd98cf6 \
- --hash=sha256:18aaacb735ab38b38db42cb01f6b92a2d0d4b6aabefeb07f02849e47f8fb3594 \
- --hash=sha256:1c832e1bd78dea67d5c16f786681b28dd695a8cb1fb90af2e27580d3d0967e92 \
- --hash=sha256:263961f658ce2165bbd7b99fa5135195c3a12d9bef045345016b8b50c315cb82 \
- --hash=sha256:271e3713e645149ea5ea3e97b57fdab61ce61333f97cfae392c28ba786f9bb49 \
- --hash=sha256:2c620736bcc334782e24d173c0fdbb7590a0a436d2fdf39310a8902505008759 \
- --hash=sha256:34716e281f181a02341ddeaad584205bd2fd3c242063bd3423d61ac259ca7eba \
- --hash=sha256:39cb9c62e471b1bb3750066ecc3a3f3052b37751c7c3dfd0fd7e48900ed52982 \
- --hash=sha256:3ac07623267feb3ae308487c260ac684b32ea35fd81e12845039952f558047b8 \
- --hash=sha256:3b0334816afb8b91dab859281b1b9786934392aa3d527cd847e41bb6f45bee65 \
- --hash=sha256:40e54d5c7e7ebf1aa596c374c49fa3135f04648a0caabcb66c52884b943f02b4 \
- --hash=sha256:50f9e62461c95d933d5c5ef4a1f2ebf9a2b4e83b0db374cb3f1de104d935922e \
- --hash=sha256:52092bc0472cfd17df49ff17e70624345efece4e1a12b23783a1ac59a1b728ed \
- --hash=sha256:5380741e53df2c566f4d234b100a484b420af85deb39ea35a1cc1be84ff53a5c \
- --hash=sha256:5e721fed53187e71d0ccf382b6bf977644c533e506c4d33c3fb24de89f5c3ed5 \
- --hash=sha256:6487aa99c2a3d509a5227d9a5e889ff05830a06b2ce08ec30df6d79db5fcd5c5 \
- --hash=sha256:6ac6310fdbfb7aa6612408bd2f07295bcbd3fda00d2d702178434751fe48e019 \
- --hash=sha256:6cfd56fc1a8e53f6e89ba3a7a7251f7396412d655bca2aa5611c8ec9a6784a1e \
- --hash=sha256:6db907c7368e3092e24919b5e31c76998b0ce1684d51a90943cb0ed1b4ffd6c1 \
- --hash=sha256:721d6b4ef5dc82ca8968c25b111e307083d7ca9091bc38163fb89243e85e3889 \
- --hash=sha256:76ad1fb5f8752eabf0fa02e4cc0336b4e8f021e2d5f061ed37d6d264db35e3ca \
- --hash=sha256:79167bba085c31f38603e11a267d862957cbb3ce018d8b38f79ac043bc92d825 \
- --hash=sha256:795c46999bae845966368a3c013e0e00947932d68e235702b5c3f6ea799aa8c9 \
- --hash=sha256:7e11270a000969409d37ed399585ee530b9ef6aa99d50c019de4cb01e8e54e62 \
- --hash=sha256:8c9ed3ba2c8a2ce098163a9bdb26f891746d02136995df25227a20e71c396ebb \
- --hash=sha256:993439ce220d25e3696d1b23b233dd010169b62f6456488567e830654ee37a6b \
- --hash=sha256:9d61e97b186a57350f6d6fd72640f9e99d5a4a2b8fbf4b9ee9a841eab327dc13 \
- --hash=sha256:9db984639887e3dffb3928d118145ffe40eff2fa40cb241a306ec57c219ebbbb \
- --hash=sha256:9e2abc762b0811e09a0d3258abee2d98e0c703eee49464ce0069590846f31d40 \
- --hash=sha256:a345928c86d535060c9c2b25e71e87c39ab2f22fc96e9636bd74d1dbf9de448c \
- --hash=sha256:ad3432cb0f9ed87477a8d97f03b763fd1d57709f1bbde3c9369b1dff5503b253 \
- --hash=sha256:ae48a786a28412d744c62fd7816a4118ef97e5be0bee968ce8f0a2fba7acf3bb \
- --hash=sha256:aef683a9ae6eb00728a542b796f52a5477b78252edede72b8327a886ab63293f \
- --hash=sha256:b90ab29d0c37ec9bf55424c064312930ca5f4bde15ee8619ee44e69319aab163 \
- --hash=sha256:c05045d8b9bfd807ee1b9f38761993297b10b245f012b11b13b91ba8945f7e45 \
- --hash=sha256:c9deabd6d547aee2c9a81dee6cc96c6d7e9a9b1953f74850c179f91fdc729cb7 \
- --hash=sha256:dde4fc32993071ac0c7dd2d82569e544f0bdaff66269cb475e0f369adad13f11 \
- --hash=sha256:eae3cf522bc7df64b42cad3925c876e1b0b6c35c1337c93e12c0f366f55b0eaf \
- --hash=sha256:ed7284b21a7a0c8f1b6e5977ac05396c0d008b89e05498c8b7e8f4a1423bba0e \
- --hash=sha256:f77f853d584e72e874d87357ad70f44b437331507d1c311457bed8ed2b956126
-six==1.17.0 \
- --hash=sha256:4721f391ed90541fddacab5acf947aa0d3dc7d27b2e1e8eda2be8970586c3274 \
- --hash=sha256:ff70335d468e7eb6ec65b95b99d3a2836546063f63acc5171de367e834932a81
-sympy==1.14.0 \
- --hash=sha256:d3d3fe8df1e5a0b42f0e7bdf50541697dbe7d23746e894990c030e2b05e72517 \
- --hash=sha256:e091cc3e99d2141a0ba2847328f5479b05d94a6635cb96148ccb3f34671bd8f5
-tabulate==0.9.0 \
- --hash=sha256:0095b12bf5966de529c0feb1fa08671671b3368eec77d7ef7ab114be2c068b3c \
- --hash=sha256:024ca478df22e9340661486f85298cff5f6dcdba14f3813e8830015b9ed1948f
-threadpoolctl==3.6.0 \
- --hash=sha256:43a0b8fd5a2928500110039e43a5eed8480b918967083ea48dc3ab9f13c4a7fb \
- --hash=sha256:8ab8b4aa3491d812b623328249fab5302a68d2d71745c8a4c719a2fcaba9f44e
-torch==2.10.0+cpu \
- --hash=sha256:0e286fcf6ce0cc7b204396c9b4ea0d375f1f0c3e752f68ce3d3aeb265511db8c \
- --hash=sha256:0fedcb1a77e8f2aaf7bfd21591bf6d1e0b207473268c9be16b17cb7783253969 \
- --hash=sha256:106dd1930cb30a4a337366ba3f9b25318ebf940f51fd46f789281dd9e736bdc4 \
- --hash=sha256:170a0623108055be5199370335cf9b41ba6875b3cb6f086db4aee583331a4899 \
- --hash=sha256:179451716487f8cb09b56459667fa1f5c4c0946c1e75fbeae77cfc40a5768d87 \
- --hash=sha256:17a09465bab2aab8f0f273410297133d8d8fb6dd84dccbd252ca4a4f3a111847 \
- --hash=sha256:1cfcb9b1558c6e52dffd0d4effce83b13c5ae5d97338164c372048c21f9cfccb \
- --hash=sha256:21cb5436978ef47c823b7a813ff0f8c2892e266cfe0f1d944879b5fba81bf4e1 \
- --hash=sha256:23882f8d882460aca809882fc42f5e343bf07585274f929ced00177d1be1eb67 \
- --hash=sha256:2adc71fe471e98a608723bfc837f7e1929885ebb912c693597711e139c1cda41 \
- --hash=sha256:31ae44836c8b9bbd1a3943d29c7c7457709ddf7c6173aa34aefe9d2203e4c405 \
- --hash=sha256:3eaa727e6a73affa61564d86b9d03191df45c8650d0666bd3d57c8597ef61e78 \
- --hash=sha256:4fcd8b4cc2ae20f2b7749fb275349c55432393868778c2d50a08e81d5ee5591e \
- --hash=sha256:5af75e5f49de21b0bdf7672bc27139bd285f9e8dbcabe2d617a2eb656514ac36 \
- --hash=sha256:6c6f0df770144907092a0d067048d96ed4f278a6c840376d2ff0e27e7579b925 \
- --hash=sha256:8d316e5bf121f1eab1147e49ad0511a9d92e4c45cc357d1ab0bee440da71a095 \
- --hash=sha256:8de5a36371b775e2d4881ed12cc7f2de400b1ad3d728aa74a281f649f87c9b8c \
- --hash=sha256:9412bd37b70f5ebd1205242c4ba4cabae35a605947f2b30806d5c9b467936db9 \
- --hash=sha256:9accc30b56cb6756d4a9d04fcb8ebc0bb68c7d55c1ed31a8657397d316d31596 \
- --hash=sha256:a280ffaea7b9c828e0c1b9b3bd502d9b6a649dc9416997b69b84544bd469f215 \
- --hash=sha256:a28fdbcfa2fbacffec81300f24dd1bed2b0ccfdbed107a823cff12bc1db070f6 \
- --hash=sha256:aada8afc068add586464b2a55adb7cc9091eec55caf5320447204741cb6a0604 \
- --hash=sha256:b67d91326e4ed9eccbd6b7d84ed7ffa43f93103aa3f0b24145f3001f3b11b714 \
- --hash=sha256:b719da5af01b59126ac13eefd6ba3dd12d002dc0e8e79b8b365e55267a8189d3 \
- --hash=sha256:b7cb1ec66cefb90fd7b676eac72cfda3b8d4e4d0cacd7a531963bc2e0a9710ab \
- --hash=sha256:ba51ef01a510baf8fff576174f702c47e1aa54389a9f1fba323bb1a5003ff0bf \
- --hash=sha256:beadc2a6a1785b09a46daad378de91ef274b8d3eea7af0bc2d017d97f115afdf \
- --hash=sha256:c35c0de592941d4944698dbfa87271ab85d3370eca3b694943a2ab307ac34b3f \
- --hash=sha256:ce5c113d1f55f8c1f5af05047a24e50d11d293e0cbbb5bf7a75c6c761edd6eaa \
- --hash=sha256:d63ee6a80982fd73fe44bb70d97d2976e010312ff6db81d7bfb9167b06dd45b9 \
- --hash=sha256:e51994492cdb76edce29da88de3672a3022f9ef0ffd90345436948d4992be2c7 \
- --hash=sha256:e71c476517c33e7db69825a9ff46c7f47a723ec4dac5b2481cff4246d1c632be \
- --hash=sha256:ea2bcc9d1fca66974a71d4bf9a502539283f35d61fcab5a799b4e120846f1e02 \
- --hash=sha256:eb1bde1ce198f05c8770017de27e001d404499cf552aaaa014569eff56ca25c0 \
- --hash=sha256:ee40b8a4b4b2cf0670c6fd4f35a7ef23871af956fecb238fbf5da15a72650b1d \
- --hash=sha256:f8294fd2fc6dd8f4435a891a0122307a043b14b21f0dac1bca63c85bfb59e586 \
- --hash=sha256:fd215f3d0f681905c5b56b0630a3d666900a37fcc3ca5b937f95275c66f9fd9c \
- --hash=sha256:ffc8da9a1341092d6a90cb5b1c1a33cd61abf0fb43f0cd88443c27fa372c26ae
-torchvision==0.25.0+cpu \
- --hash=sha256:0c2d0da9bc011a0fde1d125af396a8fbe94d99becf9d313764f24ca7657a3448 \
- --hash=sha256:2d444009c0956669ada149f61ed78f257c1cc96d259efa6acf3929ca96ceb3f0 \
- --hash=sha256:3e2ae9981e32a5b9db685659d5c7af0f04b159ff20394650a90124baf6ada51a \
- --hash=sha256:499eae1e535766391b6ee2d1e6e841239c20e2e6d88203a15b8f9f8d60a1f8bd \
- --hash=sha256:4d72a57a8f0b5146e26dac1fbfa2c905280cd04f5fcb23b9c56253506b683aeb \
- --hash=sha256:59be99d1c470ef470b134468aa6afa6f968081a503acb4ee883d70332f822e35 \
- --hash=sha256:727334e9a721cfc1ac296ce0bf9e69d9486821bfa5b1e75a8feb6f78041db481 \
- --hash=sha256:73ce04dea64914ff1110008311204c366107d651d3b04faa0dbee77efb7133b7 \
- --hash=sha256:783c8fc580bbfc159bff52f4f72cdd538e42b32956e70dffa42b940db114e151 \
- --hash=sha256:7d47d544899fabac52ebe0d4812975608fd7ab79a3d7fb6383275eb667e33f53 \
- --hash=sha256:7f851245a2687743742157988ed9c42c6e312b95bbe6cfcac9e7d0d0c28ae62f \
- --hash=sha256:813f0106eb3e268f3783da67b882458e544c6fb72f946e6ca64b5ed4e62c6a77 \
- --hash=sha256:90eec299e1f82cfaf080ccb789df3838cb9a54b57e2ebe33852cd392c692de5c \
- --hash=sha256:9212210f417888e6261c040495180f053084812cf873dedba9fc51ff4b24b2d3 \
- --hash=sha256:9511339b3b5eb75229e0b5041202e8aed9bef3b1de3a715b9fb319c9e97688fd \
- --hash=sha256:aa016ab73e06a886f72edc8929ed2ed4c85aaaa6e10500ecdef921b03129b19e \
- --hash=sha256:c1be164e93c68b2dbf460fd58975377c892dbcf3358fb72941709c3857351bba \
- --hash=sha256:c7eb5f219fdfaf1f65e68c00eb81172ab4fa08a9874dae9dad2bca360da34d0f \
- --hash=sha256:e985e12a9a232618e5a43476de5689e4b14989f5da6b93909c57afa57ec27012 \
- --hash=sha256:fb9f07f6a10f0ac24ac482ae68c6df99110b74a0d80a4c64fddc9753267d8815 \
- --hash=sha256:fe54cbd5942cd0b26a90f1748f0d4421caf67be35c281c6c3b8573733a03d630
-typing-extensions==4.15.0 \
- --hash=sha256:0cea48d173cc12fa28ecabc3b837ea3cf6f38c6d1136f85cbaaf598984861466 \
- --hash=sha256:f0fa19c6845758ab08074a0cfa8b7aecb71c999ca73d62883bc25cc018c4e548
-tzdata==2025.3 \
- --hash=sha256:06a47e5700f3081aab02b2e513160914ff0694bce9947d6b76ebd6bf57cfc5d1 \
- --hash=sha256:de39c2ca5dc7b0344f2eba86f49d614019d29f060fc4ebc8a417896a620b56a7
-vdms==0.0.23 \
- --hash=sha256:2dec153f8cb33f27cdb9ab33125198195fd29a6777cf890fc735774ed9327874 \
- --hash=sha256:7cd93242df644947c009559f31b4e19eb1dab6b62ab783ce5a2a54a4ad392f57
-wrapt==2.0.1 \
- --hash=sha256:09c7476ab884b74dce081ad9bfd07fe5822d8600abade571cb1f66d5fc915af6 \
- --hash=sha256:0e17283f533a0d24d6e5429a7d11f250a58d28b4ae5186f8f47853e3e70d2590 \
- --hash=sha256:115cae4beed3542e37866469a8a1f2b9ec549b4463572b000611e9946b86e6f6 \
- --hash=sha256:1218573502a8235bb8a7ecaed12736213b22dcde9feab115fa2989d42b5ded45 \
- --hash=sha256:17fb85fa4abc26a5184d93b3efd2dcc14deb4b09edcdb3535a536ad34f0b4dba \
- --hash=sha256:1e9b121e9aeb15df416c2c960b8255a49d44b4038016ee17af03975992d03931 \
- --hash=sha256:1f186e26ea0a55f809f232e92cc8556a0977e00183c3ebda039a807a42be1494 \
- --hash=sha256:1fdbb34da15450f2b1d735a0e969c24bdb8d8924892380126e2a293d9902078c \
- --hash=sha256:23097ed8bc4c93b7bf36fa2113c6c733c976316ce0ee2c816f64ca06102034ef \
- --hash=sha256:2879af909312d0baf35f08edeea918ee3af7ab57c37fe47cb6a373c9f2749c7b \
- --hash=sha256:2afa23318136709c4b23d87d543b425c399887b4057936cd20386d5b1422b6fa \
- --hash=sha256:2da620b31a90cdefa9cd0c2b661882329e2e19d1d7b9b920189956b76c564d75 \
- --hash=sha256:35cdbd478607036fee40273be8ed54a451f5f23121bd9d4be515158f9498f7ad \
- --hash=sha256:36982b26f190f4d737f04a492a68accbfc6fa042c3f42326fdfbb6c5b7a20a31 \
- --hash=sha256:3793ac154afb0e5b45d1233cb94d354ef7a983708cc3bb12563853b1d8d53747 \
- --hash=sha256:386fb54d9cd903ee0012c09291336469eb7b244f7183d40dc3e86a16a4bace62 \
- --hash=sha256:3cd1a4bd9a7a619922a8557e1318232e7269b5fb69d4ba97b04d20450a6bf970 \
- --hash=sha256:3d32794fe940b7000f0519904e247f902f0149edbe6316c710a8562fb6738841 \
- --hash=sha256:3d366aa598d69416b5afedf1faa539fac40c1d80a42f6b236c88c73a3c8f2d41 \
- --hash=sha256:3e271346f01e9c8b1130a6a3b0e11908049fe5be2d365a5f402778049147e7e9 \
- --hash=sha256:3f373a4ab5dbc528a94334f9fe444395b23c2f5332adab9ff4ea82f5a9e33bc1 \
- --hash=sha256:3fa272ca34332581e00bf7773e993d4f632594eb2d1b0b162a9038df0fd971dd \
- --hash=sha256:47434236c396d04875180171ee1f3815ca1eada05e24a1ee99546320d54d1d1b \
- --hash=sha256:47b0f8bafe90f7736151f61482c583c86b0693d80f075a58701dd1549b0010a9 \
- --hash=sha256:4811e15d88ee62dbf5c77f2c3ff3932b1e3ac92323ba3912f51fc4016ce81ecf \
- --hash=sha256:49989061a9977a8cbd6d20f2efa813f24bf657c6990a42967019ce779a878dbf \
- --hash=sha256:4ae879acc449caa9ed43fc36ba08392b9412ee67941748d31d94e3cedb36628c \
- --hash=sha256:4b55cacc57e1dc2d0991dbe74c6419ffd415fb66474a02335cb10efd1aa3f84f \
- --hash=sha256:4d2ce1bf1a48c5277d7969259232b57645aae5686dba1eaeade39442277afbca \
- --hash=sha256:4da7384b0e5d4cae05c97cd6f94faaf78cc8b0f791fc63af43436d98c4ab37bb \
- --hash=sha256:4e54bbf554ee29fcceee24fa41c4d091398b911da6e7f5d7bffda963c9aed2e1 \
- --hash=sha256:50844efc8cdf63b2d90cd3d62d4947a28311e6266ce5235a219d21b195b4ec2c \
- --hash=sha256:5a4939eae35db6b6cec8e7aa0e833dcca0acad8231672c26c2a9ab7a0f8ac9c8 \
- --hash=sha256:5dc1b852337c6792aa111ca8becff5bacf576bf4a0255b0f05eb749da6a1643e \
- --hash=sha256:5e53b428f65ece6d9dad23cb87e64506392b720a0b45076c05354d27a13351a1 \
- --hash=sha256:61c4956171c7434634401db448371277d07032a81cc21c599c22953374781395 \
- --hash=sha256:641e94e789b5f6b4822bb8d8ebbdfc10f4e4eae7756d648b717d980f657a9eb9 \
- --hash=sha256:64b103acdaa53b7caf409e8d45d39a8442fe6dcfec6ba3f3d141e0cc2b5b4dbd \
- --hash=sha256:68424221a2dc00d634b54f92441914929c5ffb1c30b3b837343978343a3512a3 \
- --hash=sha256:6bd1a18f5a797fe740cb3d7a0e853a8ce6461cc62023b630caec80171a6b8097 \
- --hash=sha256:6c72328f668cf4c503ffcf9434c2b71fdd624345ced7941bc6693e61bbe36bef \
- --hash=sha256:6d2d947d266d99a1477cd005b23cbd09465276e302515e122df56bb9511aca1b \
- --hash=sha256:7164a55f5e83a9a0b031d3ffab4d4e36bbec42e7025db560f225489fa929e509 \
- --hash=sha256:7b219cb2182f230676308cdcacd428fa837987b89e4b7c5c9025088b8a6c9faf \
- --hash=sha256:7d539241e87b650cbc4c3ac9f32c8d1ac8a54e510f6dca3f6ab60dcfd48c9b10 \
- --hash=sha256:7de3cc939be0e1174969f943f3b44e0d79b6f9a82198133a5b7fc6cc92882f16 \
- --hash=sha256:8330b42d769965e96e01fa14034b28a2a7600fbf7e8f0cc90ebb36d492c993e4 \
- --hash=sha256:837e31620e06b16030b1d126ed78e9383815cbac914693f54926d816d35d8edf \
- --hash=sha256:83ce30937f0ba0d28818807b303a412440c4b63e39d3d8fc036a94764b728c92 \
- --hash=sha256:85df8d92158cb8f3965aecc27cf821461bb5f40b450b03facc5d9f0d4d6ddec6 \
- --hash=sha256:8639b843c9efd84675f1e100ed9e99538ebea7297b62c4b45a7042edb84db03e \
- --hash=sha256:89a82053b193837bf93c0f8a57ded6e4b6d88033a499dadff5067e912c2a41e9 \
- --hash=sha256:8bacfe6e001749a3b64db47bcf0341da757c95959f592823a93931a422395013 \
- --hash=sha256:8ec3303e8a81932171f455f792f8df500fc1a09f20069e5c16bd7049ab4e8e38 \
- --hash=sha256:90897ea1cf0679763b62e79657958cd54eae5659f6360fc7d2ccc6f906342183 \
- --hash=sha256:908f8c6c71557f4deaa280f55d0728c3bca0960e8c3dd5ceeeafb3c19942719d \
- --hash=sha256:91bcc576260a274b169c3098e9a3519fb01f2989f6d3d386ef9cbf8653de1374 \
- --hash=sha256:9219a1d946a9b32bb23ccae66bdb61e35c62773ce7ca6509ceea70f344656b7b \
- --hash=sha256:949520bccc1fa227274da7d03bf238be15389cd94e32e4297b92337df9b7a349 \
- --hash=sha256:98d873ed6c8b4ee2418f7afce666751854d6d03e3c0ec2a399bb039cd2ae89db \
- --hash=sha256:9c9c635e78497cacb81e84f8b11b23e0aacac7a136e73b8e5b2109a1d9fc468f \
- --hash=sha256:9ca66b38dd642bf90c59b6738af8070747b610115a39af2498535f62b5cdc1c3 \
- --hash=sha256:a453257f19c31b31ba593c30d997d6e5be39e3b5ad9148c2af5a7314061c63eb \
- --hash=sha256:a52f93d95c8d38fed0669da2ebdb0b0376e895d84596a976c15a9eb45e3eccb3 \
- --hash=sha256:a9a83618c4f0757557c077ef71d708ddd9847ed66b7cc63416632af70d3e2308 \
- --hash=sha256:ab594f346517010050126fcd822697b25a7031d815bb4fbc238ccbe568216489 \
- --hash=sha256:ad3ee9d0f254851c71780966eb417ef8e72117155cff04821ab9b60549694a55 \
- --hash=sha256:aea9c7224c302bc8bfc892b908537f56c430802560e827b75ecbde81b604598b \
- --hash=sha256:b4c2e3d777e38e913b8ce3a6257af72fb608f86a1df471cb1d4339755d0a807c \
- --hash=sha256:b667189cf8efe008f55bbda321890bef628a67ab4147ebf90d182f2dadc78790 \
- --hash=sha256:b89ef9223d665ab255ae42cc282d27d69704d94be0deffc8b9d919179a609684 \
- --hash=sha256:be9e84e91d6497ba62594158d3d31ec0486c60055c49179edc51ee43d095f79c \
- --hash=sha256:bf4cb76f36be5de950ce13e22e7fdf462b35b04665a12b64f3ac5c1bbbcf3728 \
- --hash=sha256:bfb5539005259f8127ea9c885bdc231978c06b7a980e63a8a61c8c4c979719d0 \
- --hash=sha256:c046781d422f0830de6329fa4b16796096f28a92c8aef3850674442cdcb87b7f \
- --hash=sha256:c1be685ac7700c966b8610ccc63c3187a72e33cab53526a27b2a285a662cd4f7 \
- --hash=sha256:c1c91405fcf1d501fa5d55df21e58ea49e6b879ae829f1039faaf7e5e509b41e \
- --hash=sha256:c235095d6d090aa903f1db61f892fffb779c1eaeb2a50e566b52001f7a0f66ed \
- --hash=sha256:c4012a2bd37059d04f8209916aa771dfb564cccb86079072bdcd48a308b6a5c5 \
- --hash=sha256:c5ef2f2b8a53b7caee2f797ef166a390fef73979b15778a4a153e4b5fedce8fa \
- --hash=sha256:c654eafb01afac55246053d67a4b9a984a3567c3808bb7df2f8de1c1caba2e1c \
- --hash=sha256:c8d60527d1ecfc131426b10d93ab5d53e08a09c5fa0175f6b21b3252080c70a9 \
- --hash=sha256:c9e850f5b7fc67af856ff054c71690d54fa940c3ef74209ad9f935b4f66a0233 \
- --hash=sha256:cbeb0971e13b4bd81d34169ed57a6dda017328d1a22b62fda45e1d21dd06148f \
- --hash=sha256:d1a8a09a004ef100e614beec82862d11fc17d601092c3599afd22b1f36e4137e \
- --hash=sha256:d67956c676be5a24102c7407a71f4126d30de2a569a1c7871c9f3cabc94225d7 \
- --hash=sha256:d6cc985b9c8b235bd933990cdbf0f891f8e010b65a3911f7a55179cd7b0fc57b \
- --hash=sha256:d7b822c61ed04ee6ad64bc90d13368ad6eb094db54883b5dde2182f67a7f22c0 \
- --hash=sha256:df0b6d3b95932809c5b3fecc18fda0f1e07452d05e2662a0b35548985f256e28 \
- --hash=sha256:e042d653a4745be832d5aa190ff80ee4f02c34b21f4b785745eceacd0907b815 \
- --hash=sha256:e2f84e9af2060e3904a32cea9bb6db23ce3f91cfd90c6b426757cf7cc01c45c7 \
- --hash=sha256:e3612dc06b436968dfb9142c62e5dfa9eb5924f91120b3c8ff501ad878f90eb3 \
- --hash=sha256:e505629359cb5f751e16e30cf3f91a1d3ddb4552480c205947da415d597f7ac2 \
- --hash=sha256:e60690ba71a57424c8d9ff28f8d006b7ad7772c22a4af432188572cd7fa004a1 \
- --hash=sha256:e76e3f91f864e89db8b8d2a8311d57df93f01ad6bb1e9b9976d1f2e83e18315c \
- --hash=sha256:eb7cffe572ad0a141a7886a1d2efa5bef0bf7fe021deeea76b3ab334d2c38218 \
- --hash=sha256:ec65a78fbd9d6f083a15d7613b2800d5663dbb6bb96003899c834beaa68b242c \
- --hash=sha256:eda8e4ecd662d48c28bb86be9e837c13e45c58b8300e43ba3c9b4fa9900302f7 \
- --hash=sha256:f26f8e2ca19564e2e1fdbb6a0e47f36e0efbab1acc31e15471fad88f828c75f6 \
- --hash=sha256:f49027b0b9503bf6c8cdc297ca55006b80c2f5dd36cecc72c6835ab6e10e8a25 \
- --hash=sha256:f73f9f7a0ebd0db139253d27e5fc8d2866ceaeef19c30ab5d69dcbe35e1a6981 \
- --hash=sha256:fa4184e74197af3adad3c889a1af95b53bb0466bced92ea99a0c014e48323eec \
- --hash=sha256:fb1a5b72cbd751813adc02ef01ada0b0d05d3dcbc32976ce189a1279d80ad4a2 \
- --hash=sha256:fb3a86e703868561c5cad155a15c36c716e1ab513b7065bd2ac8ed353c503333 \
- --hash=sha256:fc007fdf480c77301ab1afdbb6ab22a5deee8885f3b1ed7afcb7e5e84a0e27be \
- --hash=sha256:fe21b118b9f58859b5ebaa4b130dee18669df4bd111daad082b7beb8799ad16b \
- --hash=sha256:fec0d993ecba3991645b4857837277469c8cc4c554a7e24d064d1ca291cfb81f
diff --git a/video/requirements.GPU.txt b/video/requirements.GPU.txt
deleted file mode 100644
index c0322c5..0000000
--- a/video/requirements.GPU.txt
+++ /dev/null
@@ -1,562 +0,0 @@
-colorama==0.4.6 \
- --hash=sha256:08695f5cb7ed6e0531a20572697297273c47b8cae5a63ffc6d6ed5c201be6e44 \
- --hash=sha256:4f1d9991f5acc0ca119f9d443620b77f9d6b33703e51011c16baf57afb285fc6
-coloredlogs==15.0.1 \
- --hash=sha256:612ee75c546f53e92e70049c9dbfcc18c935a2b9a53b66085ce9ef6a6e5c0934 \
- --hash=sha256:7c991aa71a4577af2f82600d8f8f3a89f936baeaf9b50a9c197da014e5bf16b0
-cuda-bindings==12.9.4 \
- --hash=sha256:1f53a7f453d4b2643d8663d036bafe29b5ba89eb904c133180f295df6dc151e5 \
- --hash=sha256:20f2699d61d724de3eb3f3369d57e2b245f93085cab44fd37c3bea036cea1a6f \
- --hash=sha256:32bdc5a76906be4c61eb98f546a6786c5773a881f3b166486449b5d141e4a39f \
- --hash=sha256:37744e721a18a514423e81863f52a4f7f46f5a6f9cccd569f2735f8067f4d8c2 \
- --hash=sha256:3adf4958dcf68ae7801a59b73fb00a8b37f8d0595060d66ceae111b1002de38d \
- --hash=sha256:443b0875916879c2e4c3722941e25e42d5ab9bcbf34c9e83404fb100fa1f6913 \
- --hash=sha256:4d3c842c2a4303b2a580fe955018e31aea30278be19795ae05226235268032e5 \
- --hash=sha256:53a10c71fdbdb743e0268d07964e5a996dd00b4e43831cbfce9804515d97d575 \
- --hash=sha256:53e11991a92ff6f26a0c8a98554cd5d6721c308a6b7bfb08bebac9201e039e43 \
- --hash=sha256:56e0043c457a99ac473ddc926fe0dc4046694d99caef633e92601ab52cbe17eb \
- --hash=sha256:694ba35023846625ef471257e6b5a4bc8af690f961d197d77d34b1d1db393f56 \
- --hash=sha256:696ca75d249ddf287d01b9a698b8e2d8a05046495a9c051ca15659dc52d17615 \
- --hash=sha256:893ca68114b5b769c1d4c02583b91ed22691887c3ed513b59467d23540104db4 \
- --hash=sha256:8b72ee72a9cc1b531db31eebaaee5c69a8ec3500e32c6933f2d3b15297b53686 \
- --hash=sha256:9866ceec83e39337d1a1d64837864c964ad902992478caa288a0bc1be95f21aa \
- --hash=sha256:a022c96b8bd847e8dc0675523431149a4c3e872f440e3002213dbb9e08f0331a \
- --hash=sha256:a2e82c8985948f953c2be51df45c3fe11c812a928fca525154fb9503190b3e64 \
- --hash=sha256:a6a429dc6c13148ff1e27c44f40a3dd23203823e637b87fd0854205195988306 \
- --hash=sha256:b32d8b685f0e66f5658bcf4601ef034e89fc2843582886f0a58784a4302da06c \
- --hash=sha256:c912a3d9e6b6651853eed8eed96d6800d69c08e94052c292fec3f282c5a817c9 \
- --hash=sha256:cf8bfaedc238f3b115d957d1fd6562b7e8435ba57f6d0e2f87d0e7149ccb2da5 \
- --hash=sha256:d80bffc357df9988dca279734bc9674c3934a654cab10cadeed27ce17d8635ee \
- --hash=sha256:f69107389e6b9948969bfd0a20c4f571fd1aefcfb1d2e1b72cc8ba5ecb7918ab \
- --hash=sha256:fda147a344e8eaeca0c6ff113d2851ffca8f7dfc0a6c932374ee5c47caa649c8
-cuda-pathfinder==1.5.1 \
- --hash=sha256:b3718097fb57cf9e8a904dd072d806f2c9a27627e35c020b06ab9454bcec08c0
-cuda-toolkit[cudart]==12.8.1 \
- --hash=sha256:adc7906af4ecbf9a352f9dca5734eceb21daec281ccfcf5675e1d2f724fc2cba
-filelock==3.25.2 \
- --hash=sha256:b64ece2b38f4ca29dd3e810287aa8c48182bbecd1ae6e9ae126c9b35f1382694 \
- --hash=sha256:ca8afb0da15f229774c9ad1b455ed96e85a81373065fb10446672f64444ddf70
-flatbuffers==25.12.19 \
- --hash=sha256:7634f50c427838bb021c2d66a3d1168e9d199b0607e6329399f04846d42e20b4
-fsspec==2026.3.0 \
- --hash=sha256:1ee6a0e28677557f8c2f994e3eea77db6392b4de9cd1f5d7a9e87a0ae9d01b41 \
- --hash=sha256:d2ceafaad1b3457968ed14efa28798162f1638dbb5d2a6868a2db002a5ee39a4
-humanfriendly==10.0 \
- --hash=sha256:1697e1a8a8f550fd43c2865cd84542fc175a61dcb779b6fee18cf6b6ccba1477 \
- --hash=sha256:6b0b831ce8f15f7300721aa49829fc4e83921a9a301cc7f606be6686a2288ddc
-jinja2==3.1.6 \
- --hash=sha256:0137fb05990d35f1275a587e9aee6d56da821fc83491a0fb838183be43f66d6d \
- --hash=sha256:85ece4451f492d0c13c5dd7c13a64681a86afae63a5f347908daf103ce6d2f67
-markupsafe==3.0.3 \
- --hash=sha256:0303439a41979d9e74d18ff5e2dd8c43ed6c6001fd40e5bf2e43f7bd9bbc523f \
- --hash=sha256:068f375c472b3e7acbe2d5318dea141359e6900156b5b2ba06a30b169086b91a \
- --hash=sha256:0bf2a864d67e76e5c9a34dc26ec616a66b9888e25e7b9460e1c76d3293bd9dbf \
- --hash=sha256:0db14f5dafddbb6d9208827849fad01f1a2609380add406671a26386cdf15a19 \
- --hash=sha256:0eb9ff8191e8498cca014656ae6b8d61f39da5f95b488805da4bb029cccbfbaf \
- --hash=sha256:0f4b68347f8c5eab4a13419215bdfd7f8c9b19f2b25520968adfad23eb0ce60c \
- --hash=sha256:1085e7fbddd3be5f89cc898938f42c0b3c711fdcb37d75221de2666af647c175 \
- --hash=sha256:116bb52f642a37c115f517494ea5feb03889e04df47eeff5b130b1808ce7c219 \
- --hash=sha256:12c63dfb4a98206f045aa9563db46507995f7ef6d83b2f68eda65c307c6829eb \
- --hash=sha256:133a43e73a802c5562be9bbcd03d090aa5a1fe899db609c29e8c8d815c5f6de6 \
- --hash=sha256:1353ef0c1b138e1907ae78e2f6c63ff67501122006b0f9abad68fda5f4ffc6ab \
- --hash=sha256:15d939a21d546304880945ca1ecb8a039db6b4dc49b2c5a400387cdae6a62e26 \
- --hash=sha256:177b5253b2834fe3678cb4a5f0059808258584c559193998be2601324fdeafb1 \
- --hash=sha256:1872df69a4de6aead3491198eaf13810b565bdbeec3ae2dc8780f14458ec73ce \
- --hash=sha256:1b4b79e8ebf6b55351f0d91fe80f893b4743f104bff22e90697db1590e47a218 \
- --hash=sha256:1b52b4fb9df4eb9ae465f8d0c228a00624de2334f216f178a995ccdcf82c4634 \
- --hash=sha256:1ba88449deb3de88bd40044603fafffb7bc2b055d626a330323a9ed736661695 \
- --hash=sha256:1cc7ea17a6824959616c525620e387f6dd30fec8cb44f649e31712db02123dad \
- --hash=sha256:218551f6df4868a8d527e3062d0fb968682fe92054e89978594c28e642c43a73 \
- --hash=sha256:26a5784ded40c9e318cfc2bdb30fe164bdb8665ded9cd64d500a34fb42067b1c \
- --hash=sha256:2713baf880df847f2bece4230d4d094280f4e67b1e813eec43b4c0e144a34ffe \
- --hash=sha256:2a15a08b17dd94c53a1da0438822d70ebcd13f8c3a95abe3a9ef9f11a94830aa \
- --hash=sha256:2f981d352f04553a7171b8e44369f2af4055f888dfb147d55e42d29e29e74559 \
- --hash=sha256:32001d6a8fc98c8cb5c947787c5d08b0a50663d139f1305bac5885d98d9b40fa \
- --hash=sha256:3524b778fe5cfb3452a09d31e7b5adefeea8c5be1d43c4f810ba09f2ceb29d37 \
- --hash=sha256:3537e01efc9d4dccdf77221fb1cb3b8e1a38d5428920e0657ce299b20324d758 \
- --hash=sha256:35add3b638a5d900e807944a078b51922212fb3dedb01633a8defc4b01a3c85f \
- --hash=sha256:38664109c14ffc9e7437e86b4dceb442b0096dfe3541d7864d9cbe1da4cf36c8 \
- --hash=sha256:3a7e8ae81ae39e62a41ec302f972ba6ae23a5c5396c8e60113e9066ef893da0d \
- --hash=sha256:3b562dd9e9ea93f13d53989d23a7e775fdfd1066c33494ff43f5418bc8c58a5c \
- --hash=sha256:457a69a9577064c05a97c41f4e65148652db078a3a509039e64d3467b9e7ef97 \
- --hash=sha256:4bd4cd07944443f5a265608cc6aab442e4f74dff8088b0dfc8238647b8f6ae9a \
- --hash=sha256:4e885a3d1efa2eadc93c894a21770e4bc67899e3543680313b09f139e149ab19 \
- --hash=sha256:4faffd047e07c38848ce017e8725090413cd80cbc23d86e55c587bf979e579c9 \
- --hash=sha256:509fa21c6deb7a7a273d629cf5ec029bc209d1a51178615ddf718f5918992ab9 \
- --hash=sha256:5678211cb9333a6468fb8d8be0305520aa073f50d17f089b5b4b477ea6e67fdc \
- --hash=sha256:591ae9f2a647529ca990bc681daebdd52c8791ff06c2bfa05b65163e28102ef2 \
- --hash=sha256:5a7d5dc5140555cf21a6fefbdbf8723f06fcd2f63ef108f2854de715e4422cb4 \
- --hash=sha256:69c0b73548bc525c8cb9a251cddf1931d1db4d2258e9599c28c07ef3580ef354 \
- --hash=sha256:6b5420a1d9450023228968e7e6a9ce57f65d148ab56d2313fcd589eee96a7a50 \
- --hash=sha256:722695808f4b6457b320fdc131280796bdceb04ab50fe1795cd540799ebe1698 \
- --hash=sha256:729586769a26dbceff69f7a7dbbf59ab6572b99d94576a5592625d5b411576b9 \
- --hash=sha256:77f0643abe7495da77fb436f50f8dab76dbc6e5fd25d39589a0f1fe6548bfa2b \
- --hash=sha256:795e7751525cae078558e679d646ae45574b47ed6e7771863fcc079a6171a0fc \
- --hash=sha256:7be7b61bb172e1ed687f1754f8e7484f1c8019780f6f6b0786e76bb01c2ae115 \
- --hash=sha256:7c3fb7d25180895632e5d3148dbdc29ea38ccb7fd210aa27acbd1201a1902c6e \
- --hash=sha256:7e68f88e5b8799aa49c85cd116c932a1ac15caaa3f5db09087854d218359e485 \
- --hash=sha256:83891d0e9fb81a825d9a6d61e3f07550ca70a076484292a70fde82c4b807286f \
- --hash=sha256:8485f406a96febb5140bfeca44a73e3ce5116b2501ac54fe953e488fb1d03b12 \
- --hash=sha256:8709b08f4a89aa7586de0aadc8da56180242ee0ada3999749b183aa23df95025 \
- --hash=sha256:8f71bc33915be5186016f675cd83a1e08523649b0e33efdb898db577ef5bb009 \
- --hash=sha256:915c04ba3851909ce68ccc2b8e2cd691618c4dc4c4232fb7982bca3f41fd8c3d \
- --hash=sha256:949b8d66bc381ee8b007cd945914c721d9aba8e27f71959d750a46f7c282b20b \
- --hash=sha256:94c6f0bb423f739146aec64595853541634bde58b2135f27f61c1ffd1cd4d16a \
- --hash=sha256:9a1abfdc021a164803f4d485104931fb8f8c1efd55bc6b748d2f5774e78b62c5 \
- --hash=sha256:9b79b7a16f7fedff2495d684f2b59b0457c3b493778c9eed31111be64d58279f \
- --hash=sha256:a320721ab5a1aba0a233739394eb907f8c8da5c98c9181d1161e77a0c8e36f2d \
- --hash=sha256:a4afe79fb3de0b7097d81da19090f4df4f8d3a2b3adaa8764138aac2e44f3af1 \
- --hash=sha256:ad2cf8aa28b8c020ab2fc8287b0f823d0a7d8630784c31e9ee5edea20f406287 \
- --hash=sha256:b8512a91625c9b3da6f127803b166b629725e68af71f8184ae7e7d54686a56d6 \
- --hash=sha256:bc51efed119bc9cfdf792cdeaa4d67e8f6fcccab66ed4bfdd6bde3e59bfcbb2f \
- --hash=sha256:bdc919ead48f234740ad807933cdf545180bfbe9342c2bb451556db2ed958581 \
- --hash=sha256:bdd37121970bfd8be76c5fb069c7751683bdf373db1ed6c010162b2a130248ed \
- --hash=sha256:be8813b57049a7dc738189df53d69395eba14fb99345e0a5994914a3864c8a4b \
- --hash=sha256:c0c0b3ade1c0b13b936d7970b1d37a57acde9199dc2aecc4c336773e1d86049c \
- --hash=sha256:c47a551199eb8eb2121d4f0f15ae0f923d31350ab9280078d1e5f12b249e0026 \
- --hash=sha256:c4ffb7ebf07cfe8931028e3e4c85f0357459a3f9f9490886198848f4fa002ec8 \
- --hash=sha256:ccfcd093f13f0f0b7fdd0f198b90053bf7b2f02a3927a30e63f3ccc9df56b676 \
- --hash=sha256:d2ee202e79d8ed691ceebae8e0486bd9a2cd4794cec4824e1c99b6f5009502f6 \
- --hash=sha256:d53197da72cc091b024dd97249dfc7794d6a56530370992a5e1a08983ad9230e \
- --hash=sha256:d6dd0be5b5b189d31db7cda48b91d7e0a9795f31430b7f271219ab30f1d3ac9d \
- --hash=sha256:d88b440e37a16e651bda4c7c2b930eb586fd15ca7406cb39e211fcff3bf3017d \
- --hash=sha256:de8a88e63464af587c950061a5e6a67d3632e36df62b986892331d4620a35c01 \
- --hash=sha256:df2449253ef108a379b8b5d6b43f4b1a8e81a061d6537becd5582fba5f9196d7 \
- --hash=sha256:e1c1493fb6e50ab01d20a22826e57520f1284df32f2d8601fdd90b6304601419 \
- --hash=sha256:e1cf1972137e83c5d4c136c43ced9ac51d0e124706ee1c8aa8532c1287fa8795 \
- --hash=sha256:e2103a929dfa2fcaf9bb4e7c091983a49c9ac3b19c9061b6d5427dd7d14d81a1 \
- --hash=sha256:e56b7d45a839a697b5eb268c82a71bd8c7f6c94d6fd50c3d577fa39a9f1409f5 \
- --hash=sha256:e8afc3f2ccfa24215f8cb28dcf43f0113ac3c37c2f0f0806d8c70e4228c5cf4d \
- --hash=sha256:e8fc20152abba6b83724d7ff268c249fa196d8259ff481f3b1476383f8f24e42 \
- --hash=sha256:eaa9599de571d72e2daf60164784109f19978b327a3910d3e9de8c97b5b70cfe \
- --hash=sha256:ec15a59cf5af7be74194f7ab02d0f59a62bdcf1a537677ce67a2537c9b87fcda \
- --hash=sha256:f190daf01f13c72eac4efd5c430a8de82489d9cff23c364c3ea822545032993e \
- --hash=sha256:f34c41761022dd093b4b6896d4810782ffbabe30f2d443ff5f083e0cbbb8c737 \
- --hash=sha256:f3e98bb3798ead92273dc0e5fd0f31ade220f59a266ffd8a4f6065e0a3ce0523 \
- --hash=sha256:f42d0984e947b8adf7dd6dde396e720934d12c506ce84eea8476409563607591 \
- --hash=sha256:f71a396b3bf33ecaa1626c255855702aca4d3d9fea5e051b41ac59a9c1c41edc \
- --hash=sha256:f9e130248f4462aaa8e2552d547f36ddadbeaa573879158d721bbd33dfe4743a \
- --hash=sha256:fed51ac40f757d41b7c48425901843666a6677e3e8eb0abcff09e4ba6e664f50
-ml-dtypes==0.5.4 \
- --hash=sha256:0d2ffd05a2575b1519dc928c0b93c06339eb67173ff53acb00724502cda231cf \
- --hash=sha256:11942cbf2cf92157db91e5022633c0d9474d4dfd813a909383bd23ce828a4b7d \
- --hash=sha256:14a4fd3228af936461db66faccef6e4f41c1d82fcc30e9f8d58a08916b1d811f \
- --hash=sha256:19b9a53598f21e453ea2fbda8aa783c20faff8e1eeb0d7ab899309a0053f1483 \
- --hash=sha256:2314892cdc3fcf05e373d76d72aaa15fda9fb98625effa73c1d646f331fcecb7 \
- --hash=sha256:2b857d3af6ac0d39db1de7c706e69c7f9791627209c3d6dedbfca8c7e5faec22 \
- --hash=sha256:304ad47faa395415b9ccbcc06a0350800bc50eda70f0e45326796e27c62f18b6 \
- --hash=sha256:35f29491a3e478407f7047b8a4834e4640a77d2737e0b294d049746507af5175 \
- --hash=sha256:388d399a2152dd79a3f0456a952284a99ee5c93d3e2f8dfe25977511e0515270 \
- --hash=sha256:3bbbe120b915090d9dd1375e4684dd17a20a2491ef25d640a908281da85e73f1 \
- --hash=sha256:3d277bf3637f2a62176f4575512e9ff9ef51d00e39626d9fe4a161992f355af2 \
- --hash=sha256:4381fe2f2452a2d7589689693d3162e876b3ddb0a832cde7a414f8e1adf7eab1 \
- --hash=sha256:4ff7f3e7ca2972e7de850e7b8fcbb355304271e2933dd90814c1cb847414d6e2 \
- --hash=sha256:531eff30e4d368cb6255bc2328d070e35836aa4f282a0fb5f3a0cd7260257298 \
- --hash=sha256:533ce891ba774eabf607172254f2e7260ba5f57bdd64030c9a4fcfbd99815d0d \
- --hash=sha256:557a31a390b7e9439056644cb80ed0735a6e3e3bb09d67fd5687e4b04238d1de \
- --hash=sha256:5a0f68ca8fd8d16583dfa7793973feb86f2fbb56ce3966daf9c9f748f52a2049 \
- --hash=sha256:6a0df4223b514d799b8a1629c65ddc351b3efa833ccf7f8ea0cf654a61d1e35d \
- --hash=sha256:6c7ecb74c4bd71db68a6bea1edf8da8c34f3d9fe218f038814fd1d310ac76c90 \
- --hash=sha256:7c23c54a00ae43edf48d44066a7ec31e05fdc2eee0be2b8b50dd1903a1db94bb \
- --hash=sha256:805cef3a38f4eafae3a5bf9ebdcdb741d0bcfd9e1bd90eb54abd24f928cd2465 \
- --hash=sha256:88c982aac7cb1cbe8cbb4e7f253072b1df872701fcaf48d84ffbb433b6568f24 \
- --hash=sha256:8ab06a50fb9bf9666dd0fe5dfb4676fa2b0ac0f31ecff72a6c3af8e22c063453 \
- --hash=sha256:8c6a2dcebd6f3903e05d51960a8058d6e131fe69f952a5397e5dbabc841b6d56 \
- --hash=sha256:8c760d85a2f82e2bed75867079188c9d18dae2ee77c25a54d60e9cc79be1bc48 \
- --hash=sha256:9ad459e99793fa6e13bd5b7e6792c8f9190b4e5a1b45c63aba14a4d0a7f1d5ff \
- --hash=sha256:9bad06436568442575beb2d03389aa7456c690a5b05892c471215bfd8cf39460 \
- --hash=sha256:a174837a64f5b16cab6f368171a1a03a27936b31699d167684073ff1c4237dac \
- --hash=sha256:a7f7c643e8b1320fd958bf098aa7ecf70623a42ec5154e3be3be673f4c34d900 \
- --hash=sha256:a9b61c19040397970d18d7737375cffd83b1f36a11dd4ad19f83a016f736c3ef \
- --hash=sha256:b4b801ebe0b477be666696bda493a9be8356f1f0057a57f1e35cd26928823e5a \
- --hash=sha256:b95e97e470fe60ed493fd9ae3911d8da4ebac16bd21f87ffa2b7c588bf22ea2c \
- --hash=sha256:bc11d7e8c44a65115d05e2ab9989d1e045125d7be8e05a071a48bc76eb6d6040 \
- --hash=sha256:bfc534409c5d4b0bf945af29e5d0ab075eae9eecbb549ff8a29280db822f34f9 \
- --hash=sha256:c1a953995cccb9e25a4ae19e34316671e4e2edaebe4cf538229b1fc7109087b7 \
- --hash=sha256:cb73dccfc991691c444acc8c0012bee8f2470da826a92e3a20bb333b1a7894e6 \
- --hash=sha256:ce756d3a10d0c4067172804c9cc276ba9cc0ff47af9078ad439b075d1abdc29b \
- --hash=sha256:d81fdb088defa30eb37bf390bb7dde35d3a83ec112ac8e33d75ab28cc29dd8b0 \
- --hash=sha256:f21c9219ef48ca5ee78402d5cc831bd58ea27ce89beda894428bc67a52da5328
-mpmath==1.3.0 \
- --hash=sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f \
- --hash=sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c
-networkx==3.4.2 \
- --hash=sha256:307c3669428c5362aab27c8a1260aa8f47c4e91d3891f48be0141738d8d053e1 \
- --hash=sha256:df5d4365b724cf81b8c6a7312509d0c22386097011ad1abe274afd5e9d3bbc5f
-numpy==2.2.6 \
- --hash=sha256:038613e9fb8c72b0a41f025a7e4c3f0b7a1b5d768ece4796b674c8f3fe13efff \
- --hash=sha256:0678000bb9ac1475cd454c6b8c799206af8107e310843532b04d49649c717a47 \
- --hash=sha256:0811bb762109d9708cca4d0b13c4f67146e3c3b7cf8d34018c722adb2d957c84 \
- --hash=sha256:0b605b275d7bd0c640cad4e5d30fa701a8d59302e127e5f79138ad62762c3e3d \
- --hash=sha256:0bca768cd85ae743b2affdc762d617eddf3bcf8724435498a1e80132d04879e6 \
- --hash=sha256:1bc23a79bfabc5d056d106f9befb8d50c31ced2fbc70eedb8155aec74a45798f \
- --hash=sha256:287cc3162b6f01463ccd86be154f284d0893d2b3ed7292439ea97eafa8170e0b \
- --hash=sha256:37c0ca431f82cd5fa716eca9506aefcabc247fb27ba69c5062a6d3ade8cf8f49 \
- --hash=sha256:37e990a01ae6ec7fe7fa1c26c55ecb672dd98b19c3d0e1d1f326fa13cb38d163 \
- --hash=sha256:389d771b1623ec92636b0786bc4ae56abafad4a4c513d36a55dce14bd9ce8571 \
- --hash=sha256:3d70692235e759f260c3d837193090014aebdf026dfd167834bcba43e30c2a42 \
- --hash=sha256:41c5a21f4a04fa86436124d388f6ed60a9343a6f767fced1a8a71c3fbca038ff \
- --hash=sha256:481b49095335f8eed42e39e8041327c05b0f6f4780488f61286ed3c01368d491 \
- --hash=sha256:4eeaae00d789f66c7a25ac5f34b71a7035bb474e679f410e5e1a94deb24cf2d4 \
- --hash=sha256:55a4d33fa519660d69614a9fad433be87e5252f4b03850642f88993f7b2ca566 \
- --hash=sha256:5a6429d4be8ca66d889b7cf70f536a397dc45ba6faeb5f8c5427935d9592e9cf \
- --hash=sha256:5bd4fc3ac8926b3819797a7c0e2631eb889b4118a9898c84f585a54d475b7e40 \
- --hash=sha256:5beb72339d9d4fa36522fc63802f469b13cdbe4fdab4a288f0c441b74272ebfd \
- --hash=sha256:6031dd6dfecc0cf9f668681a37648373bddd6421fff6c66ec1624eed0180ee06 \
- --hash=sha256:71594f7c51a18e728451bb50cc60a3ce4e6538822731b2933209a1f3614e9282 \
- --hash=sha256:74d4531beb257d2c3f4b261bfb0fc09e0f9ebb8842d82a7b4209415896adc680 \
- --hash=sha256:7befc596a7dc9da8a337f79802ee8adb30a552a94f792b9c9d18c840055907db \
- --hash=sha256:894b3a42502226a1cac872f840030665f33326fc3dac8e57c607905773cdcde3 \
- --hash=sha256:8e41fd67c52b86603a91c1a505ebaef50b3314de0213461c7a6e99c9a3beff90 \
- --hash=sha256:8e9ace4a37db23421249ed236fdcdd457d671e25146786dfc96835cd951aa7c1 \
- --hash=sha256:8fc377d995680230e83241d8a96def29f204b5782f371c532579b4f20607a289 \
- --hash=sha256:9551a499bf125c1d4f9e250377c1ee2eddd02e01eac6644c080162c0c51778ab \
- --hash=sha256:b0544343a702fa80c95ad5d3d608ea3599dd54d4632df855e4c8d24eb6ecfa1c \
- --hash=sha256:b093dd74e50a8cba3e873868d9e93a85b78e0daf2e98c6797566ad8044e8363d \
- --hash=sha256:b412caa66f72040e6d268491a59f2c43bf03eb6c96dd8f0307829feb7fa2b6fb \
- --hash=sha256:b4f13750ce79751586ae2eb824ba7e1e8dba64784086c98cdbbcc6a42112ce0d \
- --hash=sha256:b64d8d4d17135e00c8e346e0a738deb17e754230d7e0810ac5012750bbd85a5a \
- --hash=sha256:ba10f8411898fc418a521833e014a77d3ca01c15b0c6cdcce6a0d2897e6dbbdf \
- --hash=sha256:bd48227a919f1bafbdda0583705e547892342c26fb127219d60a5c36882609d1 \
- --hash=sha256:c1f9540be57940698ed329904db803cf7a402f3fc200bfe599334c9bd84a40b2 \
- --hash=sha256:c820a93b0255bc360f53eca31a0e676fd1101f673dda8da93454a12e23fc5f7a \
- --hash=sha256:ce47521a4754c8f4593837384bd3424880629f718d87c5d44f8ed763edd63543 \
- --hash=sha256:d042d24c90c41b54fd506da306759e06e568864df8ec17ccc17e9e884634fd00 \
- --hash=sha256:de749064336d37e340f640b05f24e9e3dd678c57318c7289d222a8a2f543e90c \
- --hash=sha256:e1dda9c7e08dc141e0247a5b8f49cf05984955246a327d4c48bda16821947b2f \
- --hash=sha256:e29554e2bef54a90aa5cc07da6ce955accb83f21ab5de01a62c8478897b264fd \
- --hash=sha256:e3143e4451880bed956e706a3220b4e5cf6172ef05fcc397f6f36a550b1dd868 \
- --hash=sha256:e8213002e427c69c45a52bbd94163084025f533a55a59d6f9c5b820774ef3303 \
- --hash=sha256:efd28d4e9cd7d7a8d39074a4d44c63eda73401580c5c76acda2ce969e0a38e83 \
- --hash=sha256:f0fd6321b839904e15c46e0d257fdd101dd7f530fe03fd6359c1ea63738703f3 \
- --hash=sha256:f1372f041402e37e5e633e586f62aa53de2eac8d98cbfb822806ce4bbefcb74d \
- --hash=sha256:f2618db89be1b4e05f7a1a847a9c1c0abd63e63a1607d892dd54668dd92faf87 \
- --hash=sha256:f447e6acb680fd307f40d3da4852208af94afdfab89cf850986c3ca00562f4fa \
- --hash=sha256:f92729c95468a2f4f15e9bb94c432a9229d0d50de67304399627a943201baa2f \
- --hash=sha256:f9f1adb22318e121c5c69a09142811a201ef17ab257a1e66ca3025065b7f53ae \
- --hash=sha256:fc0c5673685c508a142ca65209b4e79ed6740a4ed6b2267dbba90f34b0b3cfda \
- --hash=sha256:fc7b73d02efb0e18c000e9ad8b83480dfcd5dfd11065997ed4c6747470ae8915 \
- --hash=sha256:fd83c01228a688733f1ded5201c678f0c53ecc1006ffbc404db9f7a899ac6249 \
- --hash=sha256:fe27749d33bb772c80dcd84ae7e8df2adc920ae8297400dabec45f0dedb3f6de \
- --hash=sha256:fee4236c876c4e8369388054d02d0e9bb84821feb1a64dd59e137e6511a551f8
-nvidia-cublas-cu12==12.8.4.1 \
- --hash=sha256:47e9b82132fa8d2b4944e708049229601448aaad7e6f296f630f2d1a32de35af \
- --hash=sha256:8ac4e771d5a348c551b2a426eda6193c19aa630236b418086020df5ba9667142 \
- --hash=sha256:b86f6dd8935884615a0683b663891d43781b819ac4f2ba2b0c9604676af346d0
-nvidia-cuda-cupti-cu12==12.8.90 \
- --hash=sha256:4412396548808ddfed3f17a467b104ba7751e6b58678a4b840675c56d21cf7ed \
- --hash=sha256:bb479dcdf7e6d4f8b0b01b115260399bf34154a1a2e9fe11c85c517d87efd98e \
- --hash=sha256:ea0cb07ebda26bb9b29ba82cda34849e73c166c18162d3913575b0c9db9a6182
-nvidia-cuda-nvrtc-cu12==12.8.93 \
- --hash=sha256:7a4b6b2904850fe78e0bd179c4b655c404d4bb799ef03ddc60804247099ae909 \
- --hash=sha256:a7756528852ef889772a84c6cd89d41dfa74667e24cca16bb31f8f061e3e9994 \
- --hash=sha256:fc1fec1e1637854b4c0a65fb9a8346b51dd9ee69e61ebaccc82058441f15bce8
-nvidia-cuda-runtime-cu12==12.8.90 \
- --hash=sha256:52bf7bbee900262ffefe5e9d5a2a69a30d97e2bc5bb6cc866688caa976966e3d \
- --hash=sha256:adade8dcbd0edf427b7204d480d6066d33902cab2a4707dcfc48a2d0fd44ab90 \
- --hash=sha256:c0c6027f01505bfed6c3b21ec546f69c687689aad5f1a377554bc6ca4aa993a8
-nvidia-cudnn-cu12==9.10.2.21 \
- --hash=sha256:949452be657fa16687d0930933f032835951ef0892b37d2d53824d1a84dc97a8 \
- --hash=sha256:c6288de7d63e6cf62988f0923f96dc339cea362decb1bf5b3141883392a7d65e \
- --hash=sha256:c9132cc3f8958447b4910a1720036d9eff5928cc3179b0a51fb6d167c6cc87d8
-nvidia-cufft-cu12==11.3.3.83 \
- --hash=sha256:4d2dd21ec0b88cf61b62e6b43564355e5222e4a3fb394cac0db101f2dd0d4f74 \
- --hash=sha256:7a64a98ef2a7c47f905aaf8931b69a3a43f27c55530c698bb2ed7c75c0b42cb7 \
- --hash=sha256:848ef7224d6305cdb2a4df928759dca7b1201874787083b6e7550dd6765ce69a
-nvidia-cufile-cu12==1.13.1.3 \
- --hash=sha256:1d069003be650e131b21c932ec3d8969c1715379251f8d23a1860554b1cb24fc \
- --hash=sha256:4beb6d4cce47c1a0f1013d72e02b0994730359e17801d395bdcbf20cfb3bb00a
-nvidia-curand-cu12==10.3.9.90 \
- --hash=sha256:b32331d4f4df5d6eefa0554c565b626c7216f87a06a4f56fab27c3b68a830ec9 \
- --hash=sha256:dfab99248034673b779bc6decafdc3404a8a6f502462201f2f31f11354204acd \
- --hash=sha256:f149a8ca457277da854f89cf282d6ef43176861926c7ac85b2a0fbd237c587ec
-nvidia-cusolver-cu12==11.7.3.90 \
- --hash=sha256:4376c11ad263152bd50ea295c05370360776f8c3427b30991df774f9fb26c450 \
- --hash=sha256:4a550db115fcabc4d495eb7d39ac8b58d4ab5d8e63274d3754df1c0ad6a22d34 \
- --hash=sha256:db9ed69dbef9715071232caa9b69c52ac7de3a95773c2db65bdba85916e4e5c0
-nvidia-cusparse-cu12==12.5.8.93 \
- --hash=sha256:1ec05d76bbbd8b61b06a80e1eaf8cf4959c3d4ce8e711b65ebd0443bb0ebb13b \
- --hash=sha256:9a33604331cb2cac199f2e7f5104dfbb8a5a898c367a53dfda9ff2acb6b6b4dd \
- --hash=sha256:9b6c161cb130be1a07a27ea6923df8141f3c295852f4b260c65f18f3e0a091dc
-nvidia-cusparselt-cu12==0.7.1 \
- --hash=sha256:8878dce784d0fac90131b6817b607e803c36e629ba34dc5b433471382196b6a5 \
- --hash=sha256:f1bb701d6b930d5a7cea44c19ceb973311500847f81b634d802b7b539dc55623 \
- --hash=sha256:f67fbb5831940ec829c9117b7f33807db9f9678dc2a617fbe781cac17b4e1075
-nvidia-nccl-cu12==2.27.5 \
- --hash=sha256:31432ad4d1fb1004eb0c56203dc9bc2178a1ba69d1d9e02d64a6938ab5e40e7a \
- --hash=sha256:ad730cf15cb5d25fe849c6e6ca9eb5b76db16a80f13f425ac68d8e2e55624457
-nvidia-nvjitlink-cu12==12.8.93 \
- --hash=sha256:81ff63371a7ebd6e6451970684f916be2eab07321b73c9d244dc2b4da7f73b88 \
- --hash=sha256:adccd7161ace7261e01bb91e44e88da350895c270d23f744f0820c818b7229e7 \
- --hash=sha256:bd93fbeeee850917903583587f4fc3a4eafa022e34572251368238ab5e6bd67f
-nvidia-nvshmem-cu12==3.4.5 \
- --hash=sha256:042f2500f24c021db8a06c5eec2539027d57460e1c1a762055a6554f72c369bd \
- --hash=sha256:0b48363fc6964dede448029434c6abed6c5e37f823cb43c3bcde7ecfc0457e15
-nvidia-nvtx-cu12==12.8.90 \
- --hash=sha256:5b17e2001cc0d751a5bc2c6ec6d26ad95913324a4adb86788c944f8ce9ba441f \
- --hash=sha256:619c8304aedc69f02ea82dd244541a83c3d9d40993381b3b590f1adaed3db41e \
- --hash=sha256:d7ad891da111ebafbf7e015d34879f7112832fc239ff0d7d776b6cb685274615
-onnx==1.21.0 \
- --hash=sha256:10c3185a232089335581fabb98fba4e86d3e8246b8140f2e406082438100ebda \
- --hash=sha256:19d9971a3e52a12968ae6c70fd0f86c349536de0b0c33922ecdbe52d1972fe60 \
- --hash=sha256:1a9baf882562c4cebf79589bebb7cd71a20e30b51158cac3e3bbaf27da6163bd \
- --hash=sha256:257d1d1deb6a652913698f1e3f33ef1ca0aa69174892fe38946d4572d89dd94f \
- --hash=sha256:2aca19949260875c14866fc77ea0bc37e4e809b24976108762843d328c92d3ce \
- --hash=sha256:3abd09872523c7e0362d767e4e63bd7c6bac52a5e2c3edbf061061fe540e2027 \
- --hash=sha256:458d91948ad9a7729a347550553b49ab6939f9af2cddf334e2116e45467dc61f \
- --hash=sha256:4d8b67d0aaec5864c87633188b91cc520877477ec0254eda122bef8be43cd764 \
- --hash=sha256:5489f25fe461e7f32128218251a466cabbeeaf1eaa791c79daebf1a80d5a2cc9 \
- --hash=sha256:5f78c411743db317a76e5d009f84f7e3d5380411a1567a868e82461a1e5c775d \
- --hash=sha256:7b58a4cfec8d9311b73dc083e4c1fa362069267881144c05139b3eba5dc3a840 \
- --hash=sha256:7cd7cb8f6459311bdb557cbf6c0ccc6d8ace11c304d1bba0a30b4a4688e245f8 \
- --hash=sha256:7ee9d8fd6a4874a5fa8b44bbcabea104ce752b20469b88bc50c7dcf9030779ad \
- --hash=sha256:82aa6ab51144df07c58c4850cb78d4f1ae969d8c0bf657b28041796d49ba6974 \
- --hash=sha256:9003d5206c01fa2ff4b46311566865d8e493e1a6998d4009ec6de39843f1b59b \
- --hash=sha256:9ea4e824964082811938a9250451d89c4ec474fe42dd36c038bfa5df31993d1e \
- --hash=sha256:a9261bd580fb8548c9c37b3c6750387eb8f21ea43c63880d37b2c622e1684285 \
- --hash=sha256:ab6a488dabbb172eebc9f3b3e7ac68763f32b0c571626d4a5004608f866cc83d \
- --hash=sha256:bba12181566acf49b35875838eba49536a327b2944664b17125577d230c637ad \
- --hash=sha256:c9b56ad04039fac6b028c07e54afa1ec7f75dd340f65311f2c292e41ed7aa4d9 \
- --hash=sha256:ca14bc4842fccc3187eb538f07eabeb25a779b39388b006db4356c07403a7bbb \
- --hash=sha256:db17fc0fec46180b6acbd1d5d8650a04e5527c02b09381da0b5b888d02a204c8 \
- --hash=sha256:e0c21cc5c7a41d1a509828e2b14fe9c30e807c6df611ec0fd64a47b8d4b16abd \
- --hash=sha256:e1931bfcc222a4c9da6475f2ffffb84b97ab3876041ec639171c11ce802bee6a \
- --hash=sha256:efba467efb316baf2a9452d892c2f982b9b758c778d23e38c7f44fa211b30bb9 \
- --hash=sha256:f2c7c234c568402e10db74e33d787e4144e394ae2bcbbf11000fbfe2e017ad68 \
- --hash=sha256:f53b3c15a3b539c16b99655c43c365622046d68c49b680c48eba4da2a4fb6f27 \
- --hash=sha256:fc2635400fe39ff37ebc4e75342cc54450eadadf39c540ff132c319bf4960095
-onnxruntime-gpu==1.23.2 \
- --hash=sha256:054282614c2fc9a4a27d74242afbae706a410f1f63cc35bc72f99709029a5ba4 \
- --hash=sha256:18de50c6c8eea50acc405ea13d299aec593e46478d7a22cd32cdbbdf7c42899d \
- --hash=sha256:1e8f75af5da07329d0c3a5006087f4051d8abd133b4be7c9bae8cdab7bea4c26 \
- --hash=sha256:20959cd4ae358aab6579ab9123284a7b1498f7d51ec291d429a5edc26511306f \
- --hash=sha256:4f2d1f720685d729b5258ec1b36dee1de381b8898189908c98cbeecdb2f2b5c2 \
- --hash=sha256:7f1b3f49e5e126b99e23ec86b4203db41c2a911f6165f7624f2bc8267aaca767 \
- --hash=sha256:d76d1ac7a479ecc3ac54482eea4ba3b10d68e888a0f8b5f420f0bdf82c5eec59 \
- --hash=sha256:deba091e15357355aa836fd64c6c4ac97dd0c4609c38b08a69675073ea46b321 \
- --hash=sha256:fe925a84b00e291e0ad3fac29bfd8f8e06112abc760cdc82cb711b4f3935bd95
-onnxslim==0.1.82 \
- --hash=sha256:3190340f53c93620779f2159b41d114e571b7c1a0cfa8630cba3f7be92d3399e \
- --hash=sha256:4f48decf32863e583976fff6e9cfd9d6fe6a4a9814e7577c2cf8ce082973c6eb
-packaging==26.0 \
- --hash=sha256:00243ae351a257117b6a241061796684b084ed1c516a08c48a3f7e147a9d80b4 \
- --hash=sha256:b36f1fef9334a5588b4166f8bcd26a14e521f2b55e6b9de3aaa80d3ff7a37529
-pillow==12.2.0 \
- --hash=sha256:00a2865911330191c0b818c59103b58a5e697cae67042366970a6b6f1b20b7f9 \
- --hash=sha256:01afa7cf67f74f09523699b4e88c73fb55c13346d212a59a2db1f86b0a63e8c5 \
- --hash=sha256:03e7e372d5240cc23e9f07deca4d775c0817bffc641b01e9c3af208dbd300987 \
- --hash=sha256:03f6fab9219220f041c74aeaa2939ff0062bd5c364ba9ce037197f4c6d498cd9 \
- --hash=sha256:042db20a421b9bafecc4b84a8b6e444686bd9d836c7fd24542db3e7df7baad9b \
- --hash=sha256:0538bd5e05efec03ae613fd89c4ce0368ecd2ba239cc25b9f9be7ed426b0af1f \
- --hash=sha256:0a34329707af4f73cf1782a36cd2289c0368880654a2c11f027bcee9052d35dd \
- --hash=sha256:0c838a5125cee37e68edec915651521191cef1e6aa336b855f495766e77a366e \
- --hash=sha256:144748b3af2d1b358d41286056d0003f47cb339b8c43a9ea42f5fea4d8c66b6e \
- --hash=sha256:1610dd6c61621ae1cf811bef44d77e149ce3f7b95afe66a4512f8c59f25d9ebe \
- --hash=sha256:1e1757442ed87f4912397c6d35a0db6a7b52592156014706f17658ff58bbf795 \
- --hash=sha256:22db17c68434de69d8ecfc2fe821569195c0c373b25cccb9cbdacf2c6e53c601 \
- --hash=sha256:25373b66e0dd5905ed63fa3cae13c82fbddf3079f2c8bf15c6fb6a35586324c1 \
- --hash=sha256:2bb4a8d594eacdfc59d9e5ad972aa8afdd48d584ffd5f13a937a664c3e7db0ed \
- --hash=sha256:2c727a6d53cb0018aadd8018c2b938376af27914a68a492f59dfcaca650d5eea \
- --hash=sha256:2d192a155bbcec180f8564f693e6fd9bccff5a7af9b32e2e4bf8c9c69dbad6b5 \
- --hash=sha256:2e589959f10d9824d39b350472b92f0ce3b443c0a3442ebf41c40cb8361c5b97 \
- --hash=sha256:2e5a76d03a6c6dcef67edabda7a52494afa4035021a79c8558e14af25313d453 \
- --hash=sha256:325ca0528c6788d2a6c3d40e3568639398137346c3d6e66bb61db96b96511c98 \
- --hash=sha256:34c0d99ecccea270c04882cb3b86e7b57296079c9a4aff88cb3b33563d95afaa \
- --hash=sha256:390ede346628ccc626e5730107cde16c42d3836b89662a115a921f28440e6a3b \
- --hash=sha256:394167b21da716608eac917c60aa9b969421b5dcbbe02ae7f013e7b85811c69d \
- --hash=sha256:3997232e10d2920a68d25191392e3a4487d8183039e1c74c2297f00ed1c50705 \
- --hash=sha256:3adc9215e8be0448ed6e814966ecf3d9952f0ea40eb14e89a102b87f450660d8 \
- --hash=sha256:3e080565d8d7c671db5802eedfb438e5565ffa40115216eabb8cd52d0ecce024 \
- --hash=sha256:4a6c9fa44005fa37a91ebfc95d081e8079757d2e904b27103f4f5fa6f0bf78c0 \
- --hash=sha256:4bfd07bc812fbd20395212969e41931001fd59eb55a60658b0e5710872e95286 \
- --hash=sha256:4e6c62e9d237e9b65fac06857d511e90d8461a32adcc1b9065ea0c0fa3a28150 \
- --hash=sha256:50d8520da2a6ce0af445fa6d648c4273c3eeefbc32d7ce049f22e8b5c3daecc2 \
- --hash=sha256:51c4167c34b0d8ba05b547a3bb23578d0ba17b80a5593f93bd8ecb123dd336a3 \
- --hash=sha256:56a3f9c60a13133a98ecff6197af34d7824de9b7b38c3654861a725c970c197b \
- --hash=sha256:56b25336f502b6ed02e889f4ece894a72612fe885889a6e8c4c80239ff6e5f5f \
- --hash=sha256:57850958fe9c751670e49b2cecf6294acc99e562531f4bd317fa5ddee2068463 \
- --hash=sha256:58f62cc0f00fd29e64b29f4fd923ffdb3859c9f9e6105bfc37ba1d08994e8940 \
- --hash=sha256:5c0a9f29ca8e79f09de89293f82fc9b0270bb4af1d58bc98f540cc4aedf03166 \
- --hash=sha256:5cdfebd752ec52bf5bb4e35d9c64b40826bc5b40a13df7c3cda20a2c03a0f5ed \
- --hash=sha256:5d04bfa02cc2d23b497d1e90a0f927070043f6cbf303e738300532379a4b4e0f \
- --hash=sha256:5d2fd0fa6b5d9d1de415060363433f28da8b1526c1c129020435e186794b3795 \
- --hash=sha256:62f5409336adb0663b7caa0da5c7d9e7bdbaae9ce761d34669420c2a801b2780 \
- --hash=sha256:632ff19b2778e43162304d50da0181ce24ac5bb8180122cbe1bf4673428328c7 \
- --hash=sha256:6562ace0d3fb5f20ed7290f1f929cae41b25ae29528f2af1722966a0a02e2aa1 \
- --hash=sha256:673aa32138f3e7531ccdbca7b3901dba9b70940a19ccecc6a37c77d5fdeb05b5 \
- --hash=sha256:6a6e67ea2e6feda684ed370f9a1c52e7a243631c025ba42149a2cc5934dec295 \
- --hash=sha256:6a9adfc6d24b10f89588096364cc726174118c62130c817c2837c60cf08a392b \
- --hash=sha256:6bb77b2dcb06b20f9f4b4a8454caa581cd4dd0643a08bacf821216a16d9c8354 \
- --hash=sha256:6e6b2a0c538fc200b38ff9eb6628228b77908c319a005815f2dde585a0664b60 \
- --hash=sha256:71cde9a1e1551df7d34a25462fc60325e8a11a82cc2e2f54578e5e9a1e153d65 \
- --hash=sha256:7371b48c4fa448d20d2714c9a1f775a81155050d383333e0a6c15b1123dda005 \
- --hash=sha256:766cef22385fa1091258ad7e6216792b156dc16d8d3fa607e7545b2b72061f1c \
- --hash=sha256:7b14cc0106cd9aecda615dd6903840a058b4700fcb817687d0ee4fc8b6e389be \
- --hash=sha256:7f84204dee22a783350679a0333981df803dac21a0190d706a50475e361c93f5 \
- --hash=sha256:8023abc91fba39036dbce14a7d6535632f99c0b857807cbbbf21ecc9f4717f06 \
- --hash=sha256:80b2da48193b2f33ed0c32c38140f9d3186583ce7d516526d462645fd98660ae \
- --hash=sha256:8297651f5b5679c19968abefd6bb84d95fe30ef712eb1b2d9b2d31ca61267f4c \
- --hash=sha256:88d387ff40b3ff7c274947ed3125dedf5262ec6919d83946753b5f3d7c67ea4c \
- --hash=sha256:88ddbc66737e277852913bd1e07c150cc7bb124539f94c4e2df5344494e0a612 \
- --hash=sha256:8bd7903a5f2a4545f6fd5935c90058b89d30045568985a71c79f5fd6edf9b91e \
- --hash=sha256:8be29e59487a79f173507c30ddf57e733a357f67881430449bb32614075a40ab \
- --hash=sha256:8c984051042858021a54926eb597d6ee3012393ce9c181814115df4c60b9a808 \
- --hash=sha256:8cbeb542b2ebc6fcdacabf8aca8c1a97c9b3ad3927d46b8723f9d4f033288a0f \
- --hash=sha256:8e9c4f5b3c546fa3458a29ab22646c1c6c787ea8f5ef51300e5a60300736905e \
- --hash=sha256:90e6f81de50ad6b534cab6e5aef77ff6e37722b2f5d908686f4a5c9eba17a909 \
- --hash=sha256:975385f4776fafde056abb318f612ef6285b10a1f12b8570f3647ad0d74b48ec \
- --hash=sha256:9a8a34cc89c67a65ea7437ce257cea81a9dad65b29805f3ecee8c8fe8ff25ffe \
- --hash=sha256:9aba9a17b623ef750a4d11b742cbafffeb48a869821252b30ee21b5e91392c50 \
- --hash=sha256:9f08483a632889536b8139663db60f6724bfcb443c96f1b18855860d7d5c0fd4 \
- --hash=sha256:a4e8f36e677d3336f35089648c8955c51c6d386a13cf6ee9c189c5f5bd713a9f \
- --hash=sha256:a52edc8bfff4429aaabdf4d9ee0daadbbf8562364f940937b941f87a4290f5ff \
- --hash=sha256:a830b1a40919539d07806aa58e1b114df53ddd43213d9c8b75847eee6c0182b5 \
- --hash=sha256:aa88ccfe4e32d362816319ed727a004423aab09c5cea43c01a4b435643fa34eb \
- --hash=sha256:af73337013e0b3b46f175e79492d96845b16126ddf79c438d7ea7ff27783a414 \
- --hash=sha256:b1c1fbd8a5a1af3412a0810d060a78b5136ec0836c8a4ef9aa11807f2a22f4e1 \
- --hash=sha256:b85f66ae9eb53e860a873b858b789217ba505e5e405a24b85c0464822fe88032 \
- --hash=sha256:b86024e52a1b269467a802258c25521e6d742349d760728092e1bc2d135b4d76 \
- --hash=sha256:bd9c0c7a0c681a347b3194c500cb1e6ca9cab053ea4d82a5cf45b6b754560136 \
- --hash=sha256:bfa9c230d2fe991bed5318a5f119bd6780cda2915cca595393649fc118ab895e \
- --hash=sha256:d362d1878f00c142b7e1a16e6e5e780f02be8195123f164edf7eddd911eefe7c \
- --hash=sha256:d5d38f1411c0ed9f97bcb49b7bd59b6b7c314e0e27420e34d99d844b9ce3b6f3 \
- --hash=sha256:dac8d77255a37e81a2efcbd1fc05f1c15ee82200e6c240d7e127e25e365c39ea \
- --hash=sha256:dd025009355c926a84a612fecf58bb315a3f6814b17ead51a8e48d3823d9087f \
- --hash=sha256:deede7c263feb25dba4e82ea23058a235dcc2fe1f6021025dc71f2b618e26104 \
- --hash=sha256:e74473c875d78b8e9d5da2a70f7099549f9eb37ded4e2f6a463e60125bccd176 \
- --hash=sha256:ee3120ae9dff32f121610bb08e4313be87e03efeadfc6c0d18f89127e24d0c24 \
- --hash=sha256:eedf4b74eda2b5a4b2b2fb4c006d6295df3bf29e459e198c90ea48e130dc75c3 \
- --hash=sha256:efd8c21c98c5cc60653bcb311bef2ce0401642b7ce9d09e03a7da87c878289d4 \
- --hash=sha256:f1c943e96e85df3d3478f7b691f229887e143f81fedab9b20205349ab04d73ed \
- --hash=sha256:f278f034eb75b4e8a13a54a876cc4a5ab39173d2cdd93a638e1b467fc545ac43 \
- --hash=sha256:f3f40b3c5a968281fd507d519e444c35f0ff171237f4fdde090dd60699458421 \
- --hash=sha256:f490f9368b6fc026f021db16d7ec2fbf7d89e2edb42e8ec09d2c60505f5729c7 \
- --hash=sha256:fb043ee2f06b41473269765c2feae53fc2e2fbf96e5e22ca94fb5ad677856f06 \
- --hash=sha256:fc3d34d4a8fbec3e88a79b92e5465e0f9b842b628675850d860b8bd300b159f5
-protobuf==5.29.6 \
- --hash=sha256:36ade6ff88212e91aef4e687a971a11d7d24d6948a66751abc1b3238648f5d05 \
- --hash=sha256:62e8a3114992c7c647bce37dcc93647575fc52d50e48de30c6fcb28a6a291eb1 \
- --hash=sha256:6b9edb641441b2da9fa8f428760fc136a49cf97a52076010cf22a2ff73438a86 \
- --hash=sha256:76e07e6567f8baf827137e8d5b8204b6c7b6488bbbff1bf0a72b383f77999c18 \
- --hash=sha256:7e6ad413275be172f67fdee0f43484b6de5a904cc1c3ea9804cb6fe2ff366eda \
- --hash=sha256:831e2da16b6cc9d8f1654c041dd594eda43391affd3c03a91bea7f7f6da106d6 \
- --hash=sha256:a8866b2cff111f0f863c1b3b9e7572dc7eaea23a7fae27f6fc613304046483e6 \
- --hash=sha256:b5a169e664b4057183a34bdc424540e86eea47560f3c123a0d64de4e137f9269 \
- --hash=sha256:cb4c86de9cd8a7f3a256b9744220d87b847371c6b2f10bde87768918ef33ba49 \
- --hash=sha256:da9ee6a5424b6b30fd5e45c5ea663aef540ca95f9ad99d1e887e819cdf9b8723 \
- --hash=sha256:e3387f44798ac1106af0233c04fb8abf543772ff241169946f698b3a9a3d3ab9
-sympy==1.14.0 \
- --hash=sha256:d3d3fe8df1e5a0b42f0e7bdf50541697dbe7d23746e894990c030e2b05e72517 \
- --hash=sha256:e091cc3e99d2141a0ba2847328f5479b05d94a6635cb96148ccb3f34671bd8f5
-tensorrt-cu12==10.14.1.48.post1 \
- --hash=sha256:5a6d4d78560be7c8fff877711fa8334e8e2b441b702f047ea3107311b9897341
-tensorrt-cu12-bindings==10.14.1.48.post1 \
- --hash=sha256:03bd44f645a30e04f38d4a866bdfb0e9e16a34601384ce29ef8d008950175828 \
- --hash=sha256:1cc29c5bb32a0719d5fadbbd2e69971837b2b0d2f1575d9a2ddc1cf3f6f5d8f0 \
- --hash=sha256:2a0b5a301c84d1c95e67cede52bb49f44fda02a45c6a4b409fa16f0632394046 \
- --hash=sha256:2b83b1608e21c25da72776501533b17fe1d000595b9a191613665f67e0598868 \
- --hash=sha256:40100265c49dc91e0a3f0a030e7de0077c034e6e30c11828001b036052de1d77 \
- --hash=sha256:68ae05d23f4918fdd36a505daf8b93b64444ef9328516284363480ab776a2595 \
- --hash=sha256:78da2abb803370147e75045eaeaa2a3f134f5aa7537405f86b22eaa36c0a11ed \
- --hash=sha256:9b2c8f41d847b202e35054fbaccd55f4efbb673da11f562befc273c0b2d65f48 \
- --hash=sha256:aad15bf393acc85b2e5015f0ca2082b0d162d5966ac1f58de48d1205446e237f \
- --hash=sha256:b2bf5597c7790c36fa858b8dfd6a867a482ee41d7c35e8732b3fd671b013869c \
- --hash=sha256:b90ce26abe1d49da527211411d023f95a235806fab2d00277585558e265f9b93 \
- --hash=sha256:d9cb40e646e11225b295eaeaf74aeb7e422c425271d51ca8c416776449fec617
-tensorrt-cu12-libs==10.14.1.48.post1 \
- --hash=sha256:46e9e84e16ca7d89ca572e0900d9480945bb6faaa0c385e6f63e1ae46a834b25
-torch==2.10.0 \
- --hash=sha256:0228d20b06701c05a8f978357f657817a4a63984b0c90745def81c18aedfa591 \
- --hash=sha256:13ec4add8c3faaed8d13e0574f5cd4a323c11655546f91fbe6afa77b57423574 \
- --hash=sha256:233aed0659a2503b831d8a67e9da66a62c996204c0bba4f4c442ccc0c68a3f60 \
- --hash=sha256:29b7009dba4b7a1c960260fc8ac85022c784250af43af9fb0ebafc9883782ebd \
- --hash=sha256:2b980edd8d7c0a68c4e951ee1856334a43193f98730d97408fbd148c1a933313 \
- --hash=sha256:2c66c61f44c5f903046cc696d088e21062644cbe541c7f1c4eaae88b2ad23547 \
- --hash=sha256:3202429f58309b9fa96a614885eace4b7995729f44beb54d3e4a47773649d382 \
- --hash=sha256:3282d9febd1e4e476630a099692b44fdc214ee9bf8ee5377732d9d9dfe5712e4 \
- --hash=sha256:35e407430795c8d3edb07a1d711c41cc1f9eaddc8b2f1cc0a165a6767a8fb73d \
- --hash=sha256:418997cb02d0a0f1497cf6a09f63166f9f5df9f3e16c8a716ab76a72127c714f \
- --hash=sha256:46b3574d93a2a8134b3f5475cfb98e2eb46771794c57015f6ad1fb795ec25e49 \
- --hash=sha256:5276fa790a666ee8becaffff8acb711922252521b28fbce5db7db5cf9cb2026d \
- --hash=sha256:5c4d217b14741e40776dd7074d9006fd28b8a97ef5654db959d8635b2fe5f29b \
- --hash=sha256:5fd4117d89ffd47e3dcc71e71a22efac24828ad781c7e46aaaf56bf7f2796acf \
- --hash=sha256:6021db85958db2f07ec94e1bc77212721ba4920c12a18dc552d2ae36a3eb163f \
- --hash=sha256:6528f13d2a8593a1a412ea07a99812495bec07e9224c28b2a25c0a30c7da025c \
- --hash=sha256:682497e16bdfa6efeec8cde66531bc8d1fbbbb4d8788ec6173c089ed3cc2bfe5 \
- --hash=sha256:6b71486353fce0f9714ca0c9ef1c850a2ae766b409808acd58e9678a3edb7738 \
- --hash=sha256:6d3707a61863d1c4d6ebba7be4ca320f42b869ee657e9b2c21c736bf17000294 \
- --hash=sha256:71283a373f0ee2c89e0f0d5f446039bdabe8dbc3c9ccf35f0f784908b0acd185 \
- --hash=sha256:716b01a176c2a5659c98f6b01bf868244abdd896526f1c692712ab36dbaf9b63 \
- --hash=sha256:787124e7db3b379d4f1ed54dd12ae7c741c16a4d29b49c0226a89bea50923ffb \
- --hash=sha256:80b1b5bfe38eb0e9f5ff09f206dcac0a87aadd084230d4a36eea5ec5232c115b \
- --hash=sha256:98c01b8bb5e3240426dcde1446eed6f40c778091c8544767ef1168fc663a05a6 \
- --hash=sha256:a1ff626b884f8c4e897c4c33782bdacdff842a165fee79817b1dd549fdda1321 \
- --hash=sha256:a2f9edd8dbc99f62bc4dfb78af7bf89499bca3d753423ac1b4e06592e467b763 \
- --hash=sha256:a4be6a2a190b32ff5c8002a0977a25ea60e64f7ba46b1be37093c141d9c49aeb \
- --hash=sha256:aae1b29cd68e50a9397f5ee897b9c24742e9e306f88a807a27d617f07adb3bd8 \
- --hash=sha256:aaf663927bcd490ae971469a624c322202a2a1e68936eb952535ca4cd3b90444 \
- --hash=sha256:ac5bdcbb074384c66fa160c15b1ead77839e3fe7ed117d667249afce0acabfac \
- --hash=sha256:b1d5e2aba4eb7f8e87fbe04f86442887f9167a35f092afe4c237dfcaaef6e328 \
- --hash=sha256:b7bd80f3477b830dd166c707c5b0b82a898e7b16f59a7d9d42778dd058272e8b \
- --hash=sha256:bf0d9ff448b0218e0433aeb198805192346c4fd659c852370d5cc245f602a06a \
- --hash=sha256:c2ee399c644dc92ef7bc0d4f7e74b5360c37cdbe7c5ba11318dda49ffac2bc57 \
- --hash=sha256:cdf2a523d699b70d613243211ecaac14fe9c5df8a0b0a9c02add60fb2a413e0f \
- --hash=sha256:d8f5912ba938233f86361e891789595ff35ca4b4e2ac8fe3670895e5976731d6 \
- --hash=sha256:e521c9f030a3774ed770a9c011751fb47c4d12029a3d6522116e48431f2ff89e \
- --hash=sha256:f5ab4ba32383061be0fb74bda772d470140a12c1c3b58a0cfbf3dae94d164c28 \
- --hash=sha256:ff43db38af76fda183156153983c9a096fc4c78d0cd1e07b14a2314c7f01c2c8
-torchvision==0.25.0 \
- --hash=sha256:0b5e7f50002a8145a98c5694a018e738c50e2972608310c7e88e1bd4c058f6ce \
- --hash=sha256:0d9a3f925a081dd2ebb0b791249b687c2ef2c2717d027946654607494b9b64b6 \
- --hash=sha256:146d02c9876858420adf41f3189fe90e3d6a409cbfa65454c09f25fb33bf7266 \
- --hash=sha256:153c0d2cbc34b7cf2da19d73450f24ba36d2b75ec9211b9962b5022fb9e4ecee \
- --hash=sha256:24e11199e4d84ba9c5ee7825ebdf1cd37ce8deec225117f10243cae984ced3ec \
- --hash=sha256:40a122c3cf4d14b651f095e0f672b688dde78632783fc5cd3d4d5e4f6a828563 \
- --hash=sha256:5e6b449e9fa7d642142c0e27c41e5a43b508d57ed8e79b7c0a0c28652da8678c \
- --hash=sha256:5f271136d2d2c0b7a24c5671795c6e4fd8da4e0ea98aeb1041f62bc04c4370ef \
- --hash=sha256:620a236288d594dcec7634c754484542dc0a5c1b0e0b83a34bda5e91e9b7c3a1 \
- --hash=sha256:632db02300e83793812eee4f61ae6a2686dab10b4cfd628b620dc47747aa9d03 \
- --hash=sha256:846890161b825b38aa85fc37fb3ba5eea74e7091ff28bab378287111483b6443 \
- --hash=sha256:855c0dc6d37f462482da7531c6788518baedca1e0847f3df42a911713acdfe52 \
- --hash=sha256:a8f8061284395ce31bcd460f2169013382ccf411148ceb2ee38e718e9860f5a7 \
- --hash=sha256:a95c47abb817d4e90ea1a8e57bd0d728e3e6b533b3495ae77d84d883c4d11f56 \
- --hash=sha256:acc339aba4a858192998c2b91f635827e40d9c469d9cf1455bafdda6e4c28ea4 \
- --hash=sha256:ad9a8a5877782944d99186e4502a614770fe906626d76e9cd32446a0ac3075f2 \
- --hash=sha256:b57430fbe9e9b697418a395041bb615124d9c007710a2712fda6e35fb310f264 \
- --hash=sha256:b75deafa2dfea3e2c2a525559b04783515e3463f6e830cb71de0fb7ea36fe233 \
- --hash=sha256:c2abe430c90b1d5e552680037d68da4eb80a5852ebb1c811b2b89d299b10573b \
- --hash=sha256:c4d395cb2c4a2712f6eb93a34476cdf7aae74bb6ea2ea1917f858e96344b00aa \
- --hash=sha256:cef0196be31be421f6f462d1e9da1101be7332d91984caa6f8022e6c78a5877f \
- --hash=sha256:d1abd5ed030c708f5dbf4812ad5f6fbe9384b63c40d6bd79f8df41a4a759a917 \
- --hash=sha256:db74a551946b75d19f9996c419a799ffdf6a223ecf17c656f90da011f1d75b20 \
- --hash=sha256:ea580ffd6094cc01914ad32f8c8118174f18974629af905cea08cb6d5d48c7b7 \
- --hash=sha256:f07f01d27375ad89d72aa2b3f2180f07da95dd9d2e4c758e015c0acb2da72977 \
- --hash=sha256:f25aa9e380865b11ea6e9d99d84df86b9cc959f1a007cd966fc6f1ab2ed0e248 \
- --hash=sha256:f49964f96644dbac2506dffe1a0a7ec0f2bf8cf7a588c3319fed26e6329ffdf3 \
- --hash=sha256:f9c55ae8d673ab493325d1267cbd285bb94d56f99626c00ac4644de32a59ede3
-triton==3.6.0 \
- --hash=sha256:0b3a97e8ed304dfa9bd23bb41ca04cdf6b2e617d5e782a8653d616037a5d537d \
- --hash=sha256:10c7f76c6e72d2ef08df639e3d0d30729112f47a56b0c81672edc05ee5116ac9 \
- --hash=sha256:1722e172d34e32abc3eb7711d0025bb69d7959ebea84e3b7f7a341cd7ed694d6 \
- --hash=sha256:374f52c11a711fd062b4bfbb201fd9ac0a5febd28a96fb41b4a0f51dde3157f4 \
- --hash=sha256:448e02fe6dc898e9e5aa89cf0ee5c371e99df5aa5e8ad976a80b93334f3494fd \
- --hash=sha256:46bd1c1af4b6704e554cad2eeb3b0a6513a980d470ccfa63189737340c7746a7 \
- --hash=sha256:49df5ef37379c0c2b5c0012286f80174fcf0e073e5ade1ca9a86c36814553651 \
- --hash=sha256:6c723cfb12f6842a0ae94ac307dba7e7a44741d720a40cf0e270ed4a4e3be781 \
- --hash=sha256:74caf5e34b66d9f3a429af689c1c7128daba1d8208df60e81106b115c00d6fca \
- --hash=sha256:a17a5d5985f0ac494ed8a8e54568f092f7057ef60e1b0fa09d3fd1512064e803 \
- --hash=sha256:a6550fae429e0667e397e5de64b332d1e5695b73650ee75a6146e2e902770bea \
- --hash=sha256:d002e07d7180fd65e622134fbd980c9a3d4211fb85224b56a0a0efbd422ab72f \
- --hash=sha256:e8e323d608e3a9bfcc2d9efcc90ceefb764a82b99dea12a86d643c72539ad5d3 \
- --hash=sha256:ef5523241e7d1abca00f1d240949eebdd7c673b005edbbce0aca95b8191f1d43
-typing-extensions==4.15.0 \
- --hash=sha256:0cea48d173cc12fa28ecabc3b837ea3cf6f38c6d1136f85cbaaf598984861466 \
- --hash=sha256:f0fa19c6845758ab08074a0cfa8b7aecb71c999ca73d62883bc25cc018c4e548
-vdms==0.0.23 \
- --hash=sha256:2dec153f8cb33f27cdb9ab33125198195fd29a6777cf890fc735774ed9327874 \
- --hash=sha256:7cd93242df644947c009559f31b4e19eb1dab6b62ab783ce5a2a54a4ad392f57
-
-# The following packages are considered to be unsafe in a requirements file:
-pip==26.0.1 \
- --hash=sha256:bdb1b08f4274833d62c1aa29e20907365a2ceb950410df15fc9521bad440122b \
- --hash=sha256:c4037d8a277c89b320abe636d59f91e6d0922d08a05b60e85e53b296613346d8
diff --git a/video/requirements.txt b/video/requirements.txt
index 8784d50..5bb247f 100644
--- a/video/requirements.txt
+++ b/video/requirements.txt
@@ -1,740 +1,151 @@
build==1.4.0 \
--hash=sha256:6a07c1b8eb6f2b311b96fcbdbce5dab5fe637ffda0fd83c9cac622e927501596 \
--hash=sha256:f1b91b925aa322be454f8330c6fb48b465da993d1e7e7e6fa35027ec49f3c936
-certifi==2026.1.4 \
- --hash=sha256:9943707519e4add1115f44c2bc244f782c0249876bf51b6599fee1ffbedd685c \
- --hash=sha256:ac726dd470482006e014ad384921ed6438c457018f4b3d204aea4281258b2120
-charset-normalizer==3.4.4 \
- --hash=sha256:027f6de494925c0ab2a55eab46ae5129951638a49a34d87f4c3eda90f696b4ad \
- --hash=sha256:077fbb858e903c73f6c9db43374fd213b0b6a778106bc7032446a8e8b5b38b93 \
- --hash=sha256:0a98e6759f854bd25a58a73fa88833fba3b7c491169f86ce1180c948ab3fd394 \
- --hash=sha256:0d3d8f15c07f86e9ff82319b3d9ef6f4bf907608f53fe9d92b28ea9ae3d1fd89 \
- --hash=sha256:0f04b14ffe5fdc8c4933862d8306109a2c51e0704acfa35d51598eb45a1e89fc \
- --hash=sha256:11d694519d7f29d6cd09f6ac70028dba10f92f6cdd059096db198c283794ac86 \
- --hash=sha256:194f08cbb32dc406d6e1aea671a68be0823673db2832b38405deba2fb0d88f63 \
- --hash=sha256:1bee1e43c28aa63cb16e5c14e582580546b08e535299b8b6158a7c9c768a1f3d \
- --hash=sha256:21d142cc6c0ec30d2efee5068ca36c128a30b0f2c53c1c07bd78cb6bc1d3be5f \
- --hash=sha256:2437418e20515acec67d86e12bf70056a33abdacb5cb1655042f6538d6b085a8 \
- --hash=sha256:244bfb999c71b35de57821b8ea746b24e863398194a4014e4c76adc2bbdfeff0 \
- --hash=sha256:2677acec1a2f8ef614c6888b5b4ae4060cc184174a938ed4e8ef690e15d3e505 \
- --hash=sha256:277e970e750505ed74c832b4bf75dac7476262ee2a013f5574dd49075879e161 \
- --hash=sha256:2aaba3b0819274cc41757a1da876f810a3e4d7b6eb25699253a4effef9e8e4af \
- --hash=sha256:2b7d8f6c26245217bd2ad053761201e9f9680f8ce52f0fcd8d0755aeae5b2152 \
- --hash=sha256:2c9d3c380143a1fedbff95a312aa798578371eb29da42106a29019368a475318 \
- --hash=sha256:3162d5d8ce1bb98dd51af660f2121c55d0fa541b46dff7bb9b9f86ea1d87de72 \
- --hash=sha256:31fd66405eaf47bb62e8cd575dc621c56c668f27d46a61d975a249930dd5e2a4 \
- --hash=sha256:362d61fd13843997c1c446760ef36f240cf81d3ebf74ac62652aebaf7838561e \
- --hash=sha256:376bec83a63b8021bb5c8ea75e21c4ccb86e7e45ca4eb81146091b56599b80c3 \
- --hash=sha256:44c2a8734b333e0578090c4cd6b16f275e07aa6614ca8715e6c038e865e70576 \
- --hash=sha256:47cc91b2f4dd2833fddaedd2893006b0106129d4b94fdb6af1f4ce5a9965577c \
- --hash=sha256:4902828217069c3c5c71094537a8e623f5d097858ac6ca8252f7b4d10b7560f1 \
- --hash=sha256:4bd5d4137d500351a30687c2d3971758aac9a19208fc110ccb9d7188fbe709e8 \
- --hash=sha256:4fe7859a4e3e8457458e2ff592f15ccb02f3da787fcd31e0183879c3ad4692a1 \
- --hash=sha256:542d2cee80be6f80247095cc36c418f7bddd14f4a6de45af91dfad36d817bba2 \
- --hash=sha256:554af85e960429cf30784dd47447d5125aaa3b99a6f0683589dbd27e2f45da44 \
- --hash=sha256:5833d2c39d8896e4e19b689ffc198f08ea58116bee26dea51e362ecc7cd3ed26 \
- --hash=sha256:5947809c8a2417be3267efc979c47d76a079758166f7d43ef5ae8e9f92751f88 \
- --hash=sha256:5ae497466c7901d54b639cf42d5b8c1b6a4fead55215500d2f486d34db48d016 \
- --hash=sha256:5bd2293095d766545ec1a8f612559f6b40abc0eb18bb2f5d1171872d34036ede \
- --hash=sha256:5bfbb1b9acf3334612667b61bd3002196fe2a1eb4dd74d247e0f2a4d50ec9bbf \
- --hash=sha256:5cb4d72eea50c8868f5288b7f7f33ed276118325c1dfd3957089f6b519e1382a \
- --hash=sha256:5dbe56a36425d26d6cfb40ce79c314a2e4dd6211d51d6d2191c00bed34f354cc \
- --hash=sha256:5f819d5fe9234f9f82d75bdfa9aef3a3d72c4d24a6e57aeaebba32a704553aa0 \
- --hash=sha256:64b55f9dce520635f018f907ff1b0df1fdc31f2795a922fb49dd14fbcdf48c84 \
- --hash=sha256:6515f3182dbe4ea06ced2d9e8666d97b46ef4c75e326b79bb624110f122551db \
- --hash=sha256:65e2befcd84bc6f37095f5961e68a6f077bf44946771354a28ad434c2cce0ae1 \
- --hash=sha256:6aee717dcfead04c6eb1ce3bd29ac1e22663cdea57f943c87d1eab9a025438d7 \
- --hash=sha256:6b39f987ae8ccdf0d2642338faf2abb1862340facc796048b604ef14919e55ed \
- --hash=sha256:6e1fcf0720908f200cd21aa4e6750a48ff6ce4afe7ff5a79a90d5ed8a08296f8 \
- --hash=sha256:74018750915ee7ad843a774364e13a3db91682f26142baddf775342c3f5b1133 \
- --hash=sha256:74664978bb272435107de04e36db5a9735e78232b85b77d45cfb38f758efd33e \
- --hash=sha256:74bb723680f9f7a6234dcf67aea57e708ec1fbdf5699fb91dfd6f511b0a320ef \
- --hash=sha256:752944c7ffbfdd10c074dc58ec2d5a8a4cd9493b314d367c14d24c17684ddd14 \
- --hash=sha256:778d2e08eda00f4256d7f672ca9fef386071c9202f5e4607920b86d7803387f2 \
- --hash=sha256:780236ac706e66881f3b7f2f32dfe90507a09e67d1d454c762cf642e6e1586e0 \
- --hash=sha256:798d75d81754988d2565bff1b97ba5a44411867c0cf32b77a7e8f8d84796b10d \
- --hash=sha256:799a7a5e4fb2d5898c60b640fd4981d6a25f1c11790935a44ce38c54e985f828 \
- --hash=sha256:7a32c560861a02ff789ad905a2fe94e3f840803362c84fecf1851cb4cf3dc37f \
- --hash=sha256:7c308f7e26e4363d79df40ca5b2be1c6ba9f02bdbccfed5abddb7859a6ce72cf \
- --hash=sha256:7fa17817dc5625de8a027cb8b26d9fefa3ea28c8253929b8d6649e705d2835b6 \
- --hash=sha256:81d5eb2a312700f4ecaa977a8235b634ce853200e828fbadf3a9c50bab278328 \
- --hash=sha256:82004af6c302b5d3ab2cfc4cc5f29db16123b1a8417f2e25f9066f91d4411090 \
- --hash=sha256:837c2ce8c5a65a2035be9b3569c684358dfbf109fd3b6969630a87535495ceaa \
- --hash=sha256:840c25fb618a231545cbab0564a799f101b63b9901f2569faecd6b222ac72381 \
- --hash=sha256:8a6562c3700cce886c5be75ade4a5db4214fda19fede41d9792d100288d8f94c \
- --hash=sha256:8af65f14dc14a79b924524b1e7fffe304517b2bff5a58bf64f30b98bbc5079eb \
- --hash=sha256:8ef3c867360f88ac904fd3f5e1f902f13307af9052646963ee08ff4f131adafc \
- --hash=sha256:94537985111c35f28720e43603b8e7b43a6ecfb2ce1d3058bbe955b73404e21a \
- --hash=sha256:99ae2cffebb06e6c22bdc25801d7b30f503cc87dbd283479e7b606f70aff57ec \
- --hash=sha256:9a26f18905b8dd5d685d6d07b0cdf98a79f3c7a918906af7cc143ea2e164c8bc \
- --hash=sha256:9b35f4c90079ff2e2edc5b26c0c77925e5d2d255c42c74fdb70fb49b172726ac \
- --hash=sha256:9cd98cdc06614a2f768d2b7286d66805f94c48cde050acdbbb7db2600ab3197e \
- --hash=sha256:9d1bb833febdff5c8927f922386db610b49db6e0d4f4ee29601d71e7c2694313 \
- --hash=sha256:9f7fcd74d410a36883701fafa2482a6af2ff5ba96b9a620e9e0721e28ead5569 \
- --hash=sha256:a59cb51917aa591b1c4e6a43c132f0cdc3c76dbad6155df4e28ee626cc77a0a3 \
- --hash=sha256:a61900df84c667873b292c3de315a786dd8dac506704dea57bc957bd31e22c7d \
- --hash=sha256:a79cfe37875f822425b89a82333404539ae63dbdddf97f84dcbc3d339aae9525 \
- --hash=sha256:a8a8b89589086a25749f471e6a900d3f662d1d3b6e2e59dcecf787b1cc3a1894 \
- --hash=sha256:a8bf8d0f749c5757af2142fe7903a9df1d2e8aa3841559b2bad34b08d0e2bcf3 \
- --hash=sha256:a9768c477b9d7bd54bc0c86dbaebdec6f03306675526c9927c0e8a04e8f94af9 \
- --hash=sha256:ac1c4a689edcc530fc9d9aa11f5774b9e2f33f9a0c6a57864e90908f5208d30a \
- --hash=sha256:af2d8c67d8e573d6de5bc30cdb27e9b95e49115cd9baad5ddbd1a6207aaa82a9 \
- --hash=sha256:b435cba5f4f750aa6c0a0d92c541fb79f69a387c91e61f1795227e4ed9cece14 \
- --hash=sha256:b5b290ccc2a263e8d185130284f8501e3e36c5e02750fc6b6bdeb2e9e96f1e25 \
- --hash=sha256:b5d84d37db046c5ca74ee7bb47dd6cbc13f80665fdde3e8040bdd3fb015ecb50 \
- --hash=sha256:b7cf1017d601aa35e6bb650b6ad28652c9cd78ee6caff19f3c28d03e1c80acbf \
- --hash=sha256:bc7637e2f80d8530ee4a78e878bce464f70087ce73cf7c1caf142416923b98f1 \
- --hash=sha256:c0463276121fdee9c49b98908b3a89c39be45d86d1dbaa22957e38f6321d4ce3 \
- --hash=sha256:c4ef880e27901b6cc782f1b95f82da9313c0eb95c3af699103088fa0ac3ce9ac \
- --hash=sha256:c8ae8a0f02f57a6e61203a31428fa1d677cbe50c93622b4149d5c0f319c1d19e \
- --hash=sha256:ca5862d5b3928c4940729dacc329aa9102900382fea192fc5e52eb69d6093815 \
- --hash=sha256:cb01158d8b88ee68f15949894ccc6712278243d95f344770fa7593fa2d94410c \
- --hash=sha256:cb6254dc36b47a990e59e1068afacdcd02958bdcce30bb50cc1700a8b9d624a6 \
- --hash=sha256:cc00f04ed596e9dc0da42ed17ac5e596c6ccba999ba6bd92b0e0aef2f170f2d6 \
- --hash=sha256:cd09d08005f958f370f539f186d10aec3377d55b9eeb0d796025d4886119d76e \
- --hash=sha256:cd4b7ca9984e5e7985c12bc60a6f173f3c958eae74f3ef6624bb6b26e2abbae4 \
- --hash=sha256:ce8a0633f41a967713a59c4139d29110c07e826d131a316b50ce11b1d79b4f84 \
- --hash=sha256:cead0978fc57397645f12578bfd2d5ea9138ea0fac82b2f63f7f7c6877986a69 \
- --hash=sha256:d055ec1e26e441f6187acf818b73564e6e6282709e9bcb5b63f5b23068356a15 \
- --hash=sha256:d1f13550535ad8cff21b8d757a3257963e951d96e20ec82ab44bc64aeb62a191 \
- --hash=sha256:d9c7f57c3d666a53421049053eaacdd14bbd0a528e2186fcb2e672effd053bb0 \
- --hash=sha256:d9e45d7faa48ee908174d8fe84854479ef838fc6a705c9315372eacbc2f02897 \
- --hash=sha256:da3326d9e65ef63a817ecbcc0df6e94463713b754fe293eaa03da99befb9a5bd \
- --hash=sha256:de00632ca48df9daf77a2c65a484531649261ec9f25489917f09e455cb09ddb2 \
- --hash=sha256:e1f185f86a6f3403aa2420e815904c67b2f9ebc443f045edd0de921108345794 \
- --hash=sha256:e824f1492727fa856dd6eda4f7cee25f8518a12f3c4a56a74e8095695089cf6d \
- --hash=sha256:e912091979546adf63357d7e2ccff9b44f026c075aeaf25a52d0e95ad2281074 \
- --hash=sha256:eaabd426fe94daf8fd157c32e571c85cb12e66692f15516a83a03264b08d06c3 \
- --hash=sha256:ebf3e58c7ec8a8bed6d66a75d7fb37b55e5015b03ceae72a8e7c74495551e224 \
- --hash=sha256:ecaae4149d99b1c9e7b88bb03e3221956f68fd6d50be2ef061b2381b61d20838 \
- --hash=sha256:eecbc200c7fd5ddb9a7f16c7decb07b566c29fa2161a16cf67b8d068bd21690a \
- --hash=sha256:f155a433c2ec037d4e8df17d18922c3a0d9b3232a396690f17175d2946f0218d \
- --hash=sha256:f1e34719c6ed0b92f418c7c780480b26b5d9c50349e9a9af7d76bf757530350d \
- --hash=sha256:f34be2938726fc13801220747472850852fe6b1ea75869a048d6f896838c896f \
- --hash=sha256:f820802628d2694cb7e56db99213f930856014862f3fd943d290ea8438d07ca8 \
- --hash=sha256:f8bf04158c6b607d747e93949aa60618b61312fe647a6369f88ce2ff16043490 \
- --hash=sha256:f8e160feb2aed042cd657a72acc0b481212ed28b1b9a95c0cee1621b524e1966 \
- --hash=sha256:f9d332f8c2a2fcbffe1378594431458ddbef721c1769d78e2cbc06280d8155f9 \
- --hash=sha256:fa09f53c465e532f4d3db095e0c55b615f010ad81803d383195b6b5ca6cbf5f3 \
- --hash=sha256:faa3a41b2b66b6e50f84ae4a68c64fcd0c44355741c6374813a800cd6695db9e \
- --hash=sha256:fd44c878ea55ba351104cb93cc85e74916eb8fa440ca7903e57575e97394f608
-contourpy==1.3.2 \
- --hash=sha256:0475b1f6604896bc7c53bb070e355e9321e1bc0d381735421a2d2068ec56531f \
- --hash=sha256:106fab697af11456fcba3e352ad50effe493a90f893fca6c2ca5c033820cea92 \
- --hash=sha256:107ba8a6a7eec58bb475329e6d3b95deba9440667c4d62b9b6063942b61d7f16 \
- --hash=sha256:15ce6ab60957ca74cff444fe66d9045c1fd3e92c8936894ebd1f3eef2fff075f \
- --hash=sha256:1c48188778d4d2f3d48e4643fb15d8608b1d01e4b4d6b0548d9b336c28fc9b6f \
- --hash=sha256:3859783aefa2b8355697f16642695a5b9792e7a46ab86da1118a4a23a51a33d7 \
- --hash=sha256:3d80b2c0300583228ac98d0a927a1ba6a2ba6b8a742463c564f1d419ee5b211e \
- --hash=sha256:3f9e896f447c5c8618f1edb2bafa9a4030f22a575ec418ad70611450720b5b08 \
- --hash=sha256:434f0adf84911c924519d2b08fc10491dd282b20bdd3fa8f60fd816ea0b48841 \
- --hash=sha256:49b65a95d642d4efa8f64ba12558fcb83407e58a2dfba9d796d77b63ccfcaff5 \
- --hash=sha256:4caf2bcd2969402bf77edc4cb6034c7dd7c0803213b3523f111eb7460a51b8d2 \
- --hash=sha256:532fd26e715560721bb0d5fc7610fce279b3699b018600ab999d1be895b09415 \
- --hash=sha256:5ebac872ba09cb8f2131c46b8739a7ff71de28a24c869bcad554477eb089a878 \
- --hash=sha256:5f5964cdad279256c084b69c3f412b7801e15356b16efa9d78aa974041903da0 \
- --hash=sha256:65a887a6e8c4cd0897507d814b14c54a8c2e2aa4ac9f7686292f9769fcf9a6ab \
- --hash=sha256:6a37a2fb93d4df3fc4c0e363ea4d16f83195fc09c891bc8ce072b9d084853445 \
- --hash=sha256:70771a461aaeb335df14deb6c97439973d253ae70660ca085eec25241137ef43 \
- --hash=sha256:71e2bd4a1c4188f5c2b8d274da78faab884b59df20df63c34f74aa1813c4427c \
- --hash=sha256:745b57db7758f3ffc05a10254edd3182a2a83402a89c00957a8e8a22f5582823 \
- --hash=sha256:78e9253c3de756b3f6a5174d024c4835acd59eb3f8e2ca13e775dbffe1558f69 \
- --hash=sha256:82199cb78276249796419fe36b7386bd8d2cc3f28b3bc19fe2454fe2e26c4c15 \
- --hash=sha256:8b7fc0cd78ba2f4695fd0a6ad81a19e7e3ab825c31b577f384aa9d7817dc3bef \
- --hash=sha256:8c5acb8dddb0752bf252e01a3035b21443158910ac16a3b0d20e7fed7d534ce5 \
- --hash=sha256:8c942a01d9163e2e5cfb05cb66110121b8d07ad438a17f9e766317bcb62abf73 \
- --hash=sha256:8d2e74acbcba3bfdb6d9d8384cdc4f9260cae86ed9beee8bd5f54fee49a430b9 \
- --hash=sha256:90df94c89a91b7362e1142cbee7568f86514412ab8a2c0d0fca72d7e91b62912 \
- --hash=sha256:970e9173dbd7eba9b4e01aab19215a48ee5dd3f43cef736eebde064a171f89a5 \
- --hash=sha256:977e98a0e0480d3fe292246417239d2d45435904afd6d7332d8455981c408b85 \
- --hash=sha256:9be002b31c558d1ddf1b9b415b162c603405414bacd6932d031c5b5a8b757f0d \
- --hash=sha256:ad687a04bc802cbe8b9c399c07162a3c35e227e2daccf1668eb1f278cb698631 \
- --hash=sha256:b4f54d6a2defe9f257327b0f243612dd051cc43825587520b1bf74a31e2f6ef2 \
- --hash=sha256:b6945942715a034c671b7fc54f9588126b0b8bf23db2696e3ca8328f3ff0ab54 \
- --hash=sha256:b7cd50c38f500bbcc9b6a46643a40e0913673f869315d8e70de0438817cb7773 \
- --hash=sha256:ba38e3f9f330af820c4b27ceb4b9c7feee5fe0493ea53a8720f4792667465934 \
- --hash=sha256:c440093bbc8fc21c637c03bafcbef95ccd963bc6e0514ad887932c18ca2a759a \
- --hash=sha256:c49f73e61f1f774650a55d221803b101d966ca0c5a2d6d5e4320ec3997489441 \
- --hash=sha256:c66c4906cdbc50e9cba65978823e6e00b45682eb09adbb78c9775b74eb222422 \
- --hash=sha256:c6c4639a9c22230276b7bffb6a850dfc8258a2521305e1faefe804d006b2e532 \
- --hash=sha256:c85bb486e9be652314bb5b9e2e3b0d1b2e643d5eec4992c0fbe8ac71775da739 \
- --hash=sha256:cc829960f34ba36aad4302e78eabf3ef16a3a100863f0d4eeddf30e8a485a03b \
- --hash=sha256:cdd22595308f53ef2f891040ab2b93d79192513ffccbd7fe19be7aa773a5e09f \
- --hash=sha256:d0e589ae0d55204991450bb5c23f571c64fe43adaa53f93fc902a84c96f52fe1 \
- --hash=sha256:d14f12932a8d620e307f715857107b1d1845cc44fdb5da2bc8e850f5ceba9f87 \
- --hash=sha256:d32530b534e986374fc19eaa77fcb87e8a99e5431499949b828312bdcd20ac52 \
- --hash=sha256:d6658ccc7251a4433eebd89ed2672c2ed96fba367fd25ca9512aa92a4b46c4f1 \
- --hash=sha256:d91a3ccc7fea94ca0acab82ceb77f396d50a1f67412efe4c526f5d20264e6ecd \
- --hash=sha256:dc41ba0714aa2968d1f8674ec97504a8f7e334f48eeacebcaa6256213acb0989 \
- --hash=sha256:de39db2604ae755316cb5967728f4bea92685884b1e767b7c24e983ef5f771cb \
- --hash=sha256:de425af81b6cea33101ae95ece1f696af39446db9682a0b56daaa48cfc29f38f \
- --hash=sha256:ded1706ed0c1049224531b81128efbd5084598f18d8a2d9efae833edbd2b40ad \
- --hash=sha256:e1578f7eafce927b168752ed7e22646dad6cd9bca673c60bff55889fa236ebf9 \
- --hash=sha256:e259bced5549ac64410162adc973c5e2fb77f04df4a439d00b478e57a0e65512 \
- --hash=sha256:e298e7e70cf4eb179cc1077be1c725b5fd131ebc81181bf0c03525c8abc297fd \
- --hash=sha256:eab0f6db315fa4d70f1d8ab514e527f0366ec021ff853d7ed6a2d33605cf4b83 \
- --hash=sha256:f26b383144cf2d2c29f01a1e8170f50dacf0eac02d64139dcd709a8ac4eb3cfe \
- --hash=sha256:f939a054192ddc596e031e50bb13b657ce318cf13d264f095ce9db7dc6ae81c0 \
- --hash=sha256:fd93cc7f3139b6dd7aab2f26a90dde0aa9fc264dbf70f6740d498a70b860b82c
-cycler==0.12.1 \
- --hash=sha256:85cef7cff222d8644161529808465972e51340599459b8ac3ccbac5a854e0d30 \
- --hash=sha256:88bb128f02ba341da8ef447245a9e138fae777f6a23943da4540077d3601eb1c
-defusedxml==0.7.1 \
- --hash=sha256:1bb3032db185915b62d7c6209c5a8792be6a32ab2fedacc84e01b52c51aa3e69 \
- --hash=sha256:a352e7e428770286cc899e2542b6cdaedb2b4953ff269a210103ec58f6198a61
-filelock==3.20.3 \
- --hash=sha256:18c57ee915c7ec61cff0ecf7f0f869936c7c30191bb0cf406f1341778d0834e1 \
- --hash=sha256:4b0dda527ee31078689fc205ec4f1c1bf7d56cf88b6dc9426c4f230e46c2dce1
-fonttools==4.61.1 \
- --hash=sha256:0de30bfe7745c0d1ffa2b0b7048fb7123ad0d71107e10ee090fa0b16b9452e87 \
- --hash=sha256:10d88e55330e092940584774ee5e8a6971b01fc2f4d3466a1d6c158230880796 \
- --hash=sha256:11f35ad7805edba3aac1a3710d104592df59f4b957e30108ae0ba6c10b11dd75 \
- --hash=sha256:15acc09befd16a0fb8a8f62bc147e1a82817542d72184acca9ce6e0aeda9fa6d \
- --hash=sha256:17d2bf5d541add43822bcf0c43d7d847b160c9bb01d15d5007d84e2217aaa371 \
- --hash=sha256:2180f14c141d2f0f3da43f3a81bc8aa4684860f6b0e6f9e165a4831f24e6a23b \
- --hash=sha256:21e7c8d76f62ab13c9472ccf74515ca5b9a761d1bde3265152a6dc58700d895b \
- --hash=sha256:41a7170d042e8c0024703ed13b71893519a1a6d6e18e933e3ec7507a2c26a4b2 \
- --hash=sha256:41ed4b5ec103bd306bb68f81dc166e77409e5209443e5773cb4ed837bcc9b0d3 \
- --hash=sha256:497c31ce314219888c0e2fce5ad9178ca83fe5230b01a5006726cdf3ac9f24d9 \
- --hash=sha256:4c1b526c8d3f615a7b1867f38a9410849c8f4aef078535742198e942fba0e9bd \
- --hash=sha256:4d7092bb38c53bbc78e9255a59158b150bcdc115a1e3b3ce0b5f267dc35dd63c \
- --hash=sha256:4f5686e1fe5fce75d82d93c47a438a25bf0d1319d2843a926f741140b2b16e0c \
- --hash=sha256:58b0ee0ab5b1fc9921eccfe11d1435added19d6494dde14e323f25ad2bc30c56 \
- --hash=sha256:5ce02f38a754f207f2f06557523cd39a06438ba3aafc0639c477ac409fc64e37 \
- --hash=sha256:5fade934607a523614726119164ff621e8c30e8fa1ffffbbd358662056ba69f0 \
- --hash=sha256:5fe9fd43882620017add5eabb781ebfbc6998ee49b35bd7f8f79af1f9f99a958 \
- --hash=sha256:64102ca87e84261419c3747a0d20f396eb024bdbeb04c2bfb37e2891f5fadcb5 \
- --hash=sha256:664c5a68ec406f6b1547946683008576ef8b38275608e1cee6c061828171c118 \
- --hash=sha256:6675329885c44657f826ef01d9e4fb33b9158e9d93c537d84ad8399539bc6f69 \
- --hash=sha256:75c1a6dfac6abd407634420c93864a1e274ebc1c7531346d9254c0d8f6ca00f9 \
- --hash=sha256:75da8f28eff26defba42c52986de97b22106cb8f26515b7c22443ebc9c2d3261 \
- --hash=sha256:77efb033d8d7ff233385f30c62c7c79271c8885d5c9657d967ede124671bbdfb \
- --hash=sha256:78a7d3ab09dc47ac1a363a493e6112d8cabed7ba7caad5f54dbe2f08676d1b47 \
- --hash=sha256:7c7db70d57e5e1089a274cbb2b1fd635c9a24de809a231b154965d415d6c6d24 \
- --hash=sha256:8c56c488ab471628ff3bfa80964372fc13504ece601e0d97a78ee74126b2045c \
- --hash=sha256:91669ccac46bbc1d09e9273546181919064e8df73488ea087dcac3e2968df9ba \
- --hash=sha256:9b666a475a65f4e839d3d10473fad6d47e0a9db14a2f4a224029c5bfde58ad2c \
- --hash=sha256:9cfef3ab326780c04d6646f68d4b4742aae222e8b8ea1d627c74e38afcbc9d91 \
- --hash=sha256:a13fc8aeb24bad755eea8f7f9d409438eb94e82cf86b08fe77a03fbc8f6a96b1 \
- --hash=sha256:a75c301f96db737e1c5ed5fd7d77d9c34466de16095a266509e13da09751bd19 \
- --hash=sha256:a76d4cb80f41ba94a6691264be76435e5f72f2cb3cab0b092a6212855f71c2f6 \
- --hash=sha256:aed04cabe26f30c1647ef0e8fbb207516fd40fe9472e9439695f5c6998e60ac5 \
- --hash=sha256:b148b56f5de675ee16d45e769e69f87623a4944f7443850bf9a9376e628a89d2 \
- --hash=sha256:b501c862d4901792adaec7c25b1ecc749e2662543f68bb194c42ba18d6eec98d \
- --hash=sha256:b846a1fcf8beadeb9ea4f44ec5bdde393e2f1569e17d700bfc49cd69bde75881 \
- --hash=sha256:b931ae8f62db78861b0ff1ac017851764602288575d65b8e8ff1963fed419063 \
- --hash=sha256:c33ab3ca9d3ccd581d58e989d67554e42d8d4ded94ab3ade3508455fe70e65f7 \
- --hash=sha256:c6604b735bb12fef8e0efd5578c9fb5d3d8532d5001ea13a19cddf295673ee09 \
- --hash=sha256:d8db08051fc9e7d8bc622f2112511b8107d8f27cd89e2f64ec45e9825e8288da \
- --hash=sha256:d9203500f7c63545b4ce3799319fe4d9feb1a1b89b28d3cb5abd11b9dd64147e \
- --hash=sha256:dc492779501fa723b04d0ab1f5be046797fee17d27700476edc7ee9ae535a61e \
- --hash=sha256:e6bcdf33aec38d16508ce61fd81838f24c83c90a1d1b8c68982857038673d6b8 \
- --hash=sha256:e76ce097e3c57c4bcb67c5aa24a0ecdbd9f74ea9219997a707a4061fbe2707aa \
- --hash=sha256:eff1ac3cc66c2ac7cda1e64b4e2f3ffef474b7335f92fc3833fc632d595fcee6 \
- --hash=sha256:f3cb4a569029b9f291f88aafc927dd53683757e640081ca8c412781ea144565e \
- --hash=sha256:f79b168428351d11e10c5aeb61a74e1851ec221081299f4cf56036a95431c43a \
- --hash=sha256:fa646ecec9528bef693415c79a86e733c70a4965dd938e9a226b0fc64c9d2e6c \
- --hash=sha256:fe2efccb324948a11dd09d22136fe2ac8a97d6c1347cf0b58a911dcd529f66b7 \
- --hash=sha256:fff4f534200a04b4a36e7ae3cb74493afe807b517a09e99cb4faa89a34ed6ecd
-fsspec==2026.1.0 \
- --hash=sha256:cb76aa913c2285a3b49bdd5fc55b1d7c708d7208126b60f2eb8194fe1b4cbdcc \
- --hash=sha256:e987cb0496a0d81bba3a9d1cee62922fb395e7d4c3b575e57f547953334fe07b
+certifi==2026.2.25 \
+ --hash=sha256:027692e4402ad994f1c42e52a4997a9763c646b73e4096e4d5d6db8af1d6f0fa \
+ --hash=sha256:e887ab5cee78ea814d3472169153c2d12cd43b14bd03329a39a9c6e2e80bfba7
+charset-normalizer==3.4.6 \
+ --hash=sha256:06a7e86163334edfc5d20fe104db92fcd666e5a5df0977cb5680a506fe26cc8e \
+ --hash=sha256:0c173ce3a681f309f31b87125fecec7a5d1347261ea11ebbb856fa6006b23c8c \
+ --hash=sha256:0e28d62a8fc7a1fa411c43bd65e346f3bce9716dc51b897fbe930c5987b402d5 \
+ --hash=sha256:0e901eb1049fdb80f5bd11ed5ea1e498ec423102f7a9b9e4645d5b8204ff2815 \
+ --hash=sha256:11afb56037cbc4b1555a34dd69151e8e069bee82e613a73bef6e714ce733585f \
+ --hash=sha256:150b8ce8e830eb7ccb029ec9ca36022f756986aaaa7956aad6d9ec90089338c0 \
+ --hash=sha256:172985e4ff804a7ad08eebec0a1640ece87ba5041d565fff23c8f99c1f389484 \
+ --hash=sha256:197c1a244a274bb016dd8b79204850144ef77fe81c5b797dc389327adb552407 \
+ --hash=sha256:1ae6b62897110aa7c79ea2f5dd38d1abca6db663687c0b1ad9aed6f6bae3d9d6 \
+ --hash=sha256:1cf0a70018692f85172348fe06d3a4b63f94ecb055e13a00c644d368eb82e5b8 \
+ --hash=sha256:1ed80ff870ca6de33f4d953fda4d55654b9a2b340ff39ab32fa3adbcd718f264 \
+ --hash=sha256:22c6f0c2fbc31e76c3b8a86fba1a56eda6166e238c29cdd3d14befdb4a4e4815 \
+ --hash=sha256:231d4da14bcd9301310faf492051bee27df11f2bc7549bc0bb41fef11b82daa2 \
+ --hash=sha256:259695e2ccc253feb2a016303543d691825e920917e31f894ca1a687982b1de4 \
+ --hash=sha256:2a24157fa36980478dd1770b585c0f30d19e18f4fb0c47c13aa568f871718579 \
+ --hash=sha256:2b1a63e8224e401cafe7739f77efd3f9e7f5f2026bda4aead8e59afab537784f \
+ --hash=sha256:2bd9d128ef93637a5d7a6af25363cf5dec3fa21cf80e68055aad627f280e8afa \
+ --hash=sha256:2e1d8ca8611099001949d1cdfaefc510cf0f212484fe7c565f735b68c78c3c95 \
+ --hash=sha256:2ef7fedc7a6ecbe99969cd09632516738a97eeb8bd7258bf8a0f23114c057dab \
+ --hash=sha256:2f7fdd9b6e6c529d6a2501a2d36b240109e78a8ceaef5687cfcfa2bbe671d297 \
+ --hash=sha256:30f445ae60aad5e1f8bdbb3108e39f6fbc09f4ea16c815c66578878325f8f15a \
+ --hash=sha256:31215157227939b4fb3d740cd23fe27be0439afef67b785a1eb78a3ae69cba9e \
+ --hash=sha256:34315ff4fc374b285ad7f4a0bf7dcbfe769e1b104230d40f49f700d4ab6bbd84 \
+ --hash=sha256:3516bbb8d42169de9e61b8520cbeeeb716f12f4ecfe3fd30a9919aa16c806ca8 \
+ --hash=sha256:3778fd7d7cd04ae8f54651f4a7a0bd6e39a0cf20f801720a4c21d80e9b7ad6b0 \
+ --hash=sha256:39f5068d35621da2881271e5c3205125cc456f54e9030d3f723288c873a71bf9 \
+ --hash=sha256:404a1e552cf5b675a87f0651f8b79f5f1e6fd100ee88dc612f89aa16abd4486f \
+ --hash=sha256:419a9d91bd238052642a51938af8ac05da5b3343becde08d5cdeab9046df9ee1 \
+ --hash=sha256:423fb7e748a08f854a08a222b983f4df1912b1daedce51a72bd24fe8f26a1843 \
+ --hash=sha256:4482481cb0572180b6fd976a4d5c72a30263e98564da68b86ec91f0fe35e8565 \
+ --hash=sha256:461598cd852bfa5a61b09cae2b1c02e2efcd166ee5516e243d540ac24bfa68a7 \
+ --hash=sha256:47955475ac79cc504ef2704b192364e51d0d473ad452caedd0002605f780101c \
+ --hash=sha256:48696db7f18afb80a068821504296eb0787d9ce239b91ca15059d1d3eaacf13b \
+ --hash=sha256:4be9f4830ba8741527693848403e2c457c16e499100963ec711b1c6f2049b7c7 \
+ --hash=sha256:4d1d02209e06550bdaef34af58e041ad71b88e624f5d825519da3a3308e22687 \
+ --hash=sha256:4f41da960b196ea355357285ad1316a00099f22d0929fe168343b99b254729c9 \
+ --hash=sha256:517ad0e93394ac532745129ceabdf2696b609ec9f87863d337140317ebce1c14 \
+ --hash=sha256:51fb3c322c81d20567019778cb5a4a6f2dc1c200b886bc0d636238e364848c89 \
+ --hash=sha256:5273b9f0b5835ff0350c0828faea623c68bfa65b792720c453e22b25cc72930f \
+ --hash=sha256:530d548084c4a9f7a16ed4a294d459b4f229db50df689bfe92027452452943a0 \
+ --hash=sha256:530e8cebeea0d76bdcf93357aa5e41336f48c3dc709ac52da2bb167c5b8271d9 \
+ --hash=sha256:54fae94be3d75f3e573c9a1b5402dc593de19377013c9a0e4285e3d402dd3a2a \
+ --hash=sha256:572d7c822caf521f0525ba1bce1a622a0b85cf47ffbdae6c9c19e3b5ac3c4389 \
+ --hash=sha256:58c948d0d086229efc484fe2f30c2d382c86720f55cd9bc33591774348ad44e0 \
+ --hash=sha256:5d11595abf8dd942a77883a39d81433739b287b6aa71620f15164f8096221b30 \
+ --hash=sha256:5f8ddd609f9e1af8c7bd6e2aca279c931aefecd148a14402d4e368f3171769fd \
+ --hash=sha256:5feb91325bbceade6afab43eb3b508c63ee53579fe896c77137ded51c6b6958e \
+ --hash=sha256:60c74963d8350241a79cb8feea80e54d518f72c26db618862a8f53e5023deaf9 \
+ --hash=sha256:613f19aa6e082cf96e17e3ffd89383343d0d589abda756b7764cf78361fd41dc \
+ --hash=sha256:659a1e1b500fac8f2779dd9e1570464e012f43e580371470b45277a27baa7532 \
+ --hash=sha256:695f5c2823691a25f17bc5d5ffe79fa90972cc34b002ac6c843bb8a1720e950d \
+ --hash=sha256:69dd852c2f0ad631b8b60cfbe25a28c0058a894de5abb566619c205ce0550eae \
+ --hash=sha256:6cceb5473417d28edd20c6c984ab6fee6c6267d38d906823ebfe20b03d607dc2 \
+ --hash=sha256:71be7e0e01753a89cf024abf7ecb6bca2c81738ead80d43004d9b5e3f1244e64 \
+ --hash=sha256:74119174722c4349af9708993118581686f343adc1c8c9c007d59be90d077f3f \
+ --hash=sha256:74a2e659c7ecbc73562e2a15e05039f1e22c75b7c7618b4b574a3ea9118d1557 \
+ --hash=sha256:7504e9b7dc05f99a9bbb4525c67a2c155073b44d720470a148b34166a69c054e \
+ --hash=sha256:79090741d842f564b1b2827c0b82d846405b744d31e84f18d7a7b41c20e473ff \
+ --hash=sha256:7a6967aaf043bceabab5412ed6bd6bd26603dae84d5cb75bf8d9a74a4959d398 \
+ --hash=sha256:7bda6eebafd42133efdca535b04ccb338ab29467b3f7bf79569883676fc628db \
+ --hash=sha256:7edbed096e4a4798710ed6bc75dcaa2a21b68b6c356553ac4823c3658d53743a \
+ --hash=sha256:7f9019c9cb613f084481bd6a100b12e1547cf2efe362d873c2e31e4035a6fa43 \
+ --hash=sha256:802168e03fba8bbc5ce0d866d589e4b1ca751d06edee69f7f3a19c5a9fe6b597 \
+ --hash=sha256:80d0a5615143c0b3225e5e3ef22c8d5d51f3f72ce0ea6fb84c943546c7b25b6c \
+ --hash=sha256:82060f995ab5003a2d6e0f4ad29065b7672b6593c8c63559beefe5b443242c3e \
+ --hash=sha256:836ab36280f21fc1a03c99cd05c6b7af70d2697e374c7af0b61ed271401a72a2 \
+ --hash=sha256:8761ac29b6c81574724322a554605608a9960769ea83d2c73e396f3df896ad54 \
+ --hash=sha256:87725cfb1a4f1f8c2fc9890ae2f42094120f4b44db9360be5d99a4c6b0e03a9e \
+ --hash=sha256:899d28f422116b08be5118ef350c292b36fc15ec2daeb9ea987c89281c7bb5c4 \
+ --hash=sha256:8bc5f0687d796c05b1e28ab0d38a50e6309906ee09375dd3aff6a9c09dd6e8f4 \
+ --hash=sha256:8bea55c4eef25b0b19a0337dc4e3f9a15b00d569c77211fa8cde38684f234fb7 \
+ --hash=sha256:8e5a94886bedca0f9b78fecd6afb6629142fd2605aa70a125d49f4edc6037ee6 \
+ --hash=sha256:90ca27cd8da8118b18a52d5f547859cc1f8354a00cd1e8e5120df3e30d6279e5 \
+ --hash=sha256:92734d4d8d187a354a556626c221cd1a892a4e0802ccb2af432a1d85ec012194 \
+ --hash=sha256:947cf925bc916d90adba35a64c82aace04fa39b46b52d4630ece166655905a69 \
+ --hash=sha256:95b52c68d64c1878818687a473a10547b3292e82b6f6fe483808fb1468e2f52f \
+ --hash=sha256:97d0235baafca5f2b09cf332cc275f021e694e8362c6bb9c96fc9a0eb74fc316 \
+ --hash=sha256:9ca4c0b502ab399ef89248a2c84c54954f77a070f28e546a85e91da627d1301e \
+ --hash=sha256:9cc4fc6c196d6a8b76629a70ddfcd4635a6898756e2d9cac5565cf0654605d73 \
+ --hash=sha256:9cc6e6d9e571d2f863fa77700701dae73ed5f78881efc8b3f9a4398772ff53e8 \
+ --hash=sha256:a056d1ad2633548ca18ffa2f85c202cfb48b68615129143915b8dc72a806a923 \
+ --hash=sha256:a26611d9987b230566f24a0a125f17fe0de6a6aff9f25c9f564aaa2721a5fb88 \
+ --hash=sha256:a4474d924a47185a06411e0064b803c68be044be2d60e50e8bddcc2649957c1f \
+ --hash=sha256:a4ea868bc28109052790eb2b52a9ab33f3aa7adc02f96673526ff47419490e21 \
+ --hash=sha256:a9e68c9d88823b274cf1e72f28cb5dc89c990edf430b0bfd3e2fb0785bfeabf4 \
+ --hash=sha256:aa9cccf4a44b9b62d8ba8b4dd06c649ba683e4bf04eea606d2e94cfc2d6ff4d6 \
+ --hash=sha256:ab30e5e3e706e3063bc6de96b118688cb10396b70bb9864a430f67df98c61ecc \
+ --hash=sha256:ac2393c73378fea4e52aa56285a3d64be50f1a12395afef9cce47772f60334c2 \
+ --hash=sha256:ad8faf8df23f0378c6d527d8b0b15ea4a2e23c89376877c598c4870d1b2c7866 \
+ --hash=sha256:b35b200d6a71b9839a46b9b7fff66b6638bb52fc9658aa58796b0326595d3021 \
+ --hash=sha256:b3694e3f87f8ac7ce279d4355645b3c878d24d1424581b46282f24b92f5a4ae2 \
+ --hash=sha256:b4ff1d35e8c5bd078be89349b6f3a845128e685e751b6ea1169cf2160b344c4d \
+ --hash=sha256:bbc8c8650c6e51041ad1be191742b8b421d05bbd3410f43fa2a00c8db87678e8 \
+ --hash=sha256:bc72863f4d9aba2e8fd9085e63548a324ba706d2ea2c83b260da08a59b9482de \
+ --hash=sha256:bf625105bb9eef28a56a943fec8c8a98aeb80e7d7db99bd3c388137e6eb2d237 \
+ --hash=sha256:c2274ca724536f173122f36c98ce188fd24ce3dad886ec2b7af859518ce008a4 \
+ --hash=sha256:c45a03a4c69820a399f1dda9e1d8fbf3562eda46e7720458180302021b08f778 \
+ --hash=sha256:c8ae56368f8cc97c7e40a7ee18e1cedaf8e780cd8bc5ed5ac8b81f238614facb \
+ --hash=sha256:c907cdc8109f6c619e6254212e794d6548373cc40e1ec75e6e3823d9135d29cc \
+ --hash=sha256:ca0276464d148c72defa8bb4390cce01b4a0e425f3b50d1435aa6d7a18107602 \
+ --hash=sha256:cd5e2801c89992ed8c0a3f0293ae83c159a60d9a5d685005383ef4caca77f2c4 \
+ --hash=sha256:d08ec48f0a1c48d75d0356cea971921848fb620fdeba805b28f937e90691209f \
+ --hash=sha256:d1a2ee9c1499fc8f86f4521f27a973c914b211ffa87322f4ee33bb35392da2c5 \
+ --hash=sha256:d5f5d1e9def3405f60e3ca8232d56f35c98fb7bf581efcc60051ebf53cb8b611 \
+ --hash=sha256:d60377dce4511655582e300dc1e5a5f24ba0cb229005a1d5c8d0cb72bb758ab8 \
+ --hash=sha256:d73beaac5e90173ac3deb9928a74763a6d230f494e4bfb422c217a0ad8e629bf \
+ --hash=sha256:d7de2637729c67d67cf87614b566626057e95c303bc0a55ffe391f5205e7003d \
+ --hash=sha256:dad6e0f2e481fffdcf776d10ebee25e0ef89f16d691f1e5dee4b586375fdc64b \
+ --hash=sha256:dda86aba335c902b6149a02a55b38e96287157e609200811837678214ba2b1db \
+ --hash=sha256:df01808ee470038c3f8dc4f48620df7225c49c2d6639e38f96e6d6ac6e6f7b0e \
+ --hash=sha256:e1f6e2f00a6b8edb562826e4632e26d063ac10307e80f7461f7de3ad8ef3f077 \
+ --hash=sha256:e25369dc110d58ddf29b949377a93e0716d72a24f62bad72b2b39f155949c1fd \
+ --hash=sha256:e3c701e954abf6fc03a49f7c579cc80c2c6cc52525340ca3186c41d3f33482ef \
+ --hash=sha256:e5bcc1a1ae744e0bb59641171ae53743760130600da8db48cbb6e4918e186e4e \
+ --hash=sha256:e68c14b04827dd76dcbd1aeea9e604e3e4b78322d8faf2f8132c7138efa340a8 \
+ --hash=sha256:e8aeb10fcbe92767f0fa69ad5a72deca50d0dca07fbde97848997d778a50c9fe \
+ --hash=sha256:e985a16ff513596f217cee86c21371b8cd011c0f6f056d0920aa2d926c544058 \
+ --hash=sha256:ecbbd45615a6885fe3240eb9db73b9e62518b611850fdf8ab08bd56de7ad2b17 \
+ --hash=sha256:ee4ec14bc1680d6b0afab9aea2ef27e26d2024f18b24a2d7155a52b60da7e833 \
+ --hash=sha256:ef5960d965e67165d75b7c7ffc60a83ec5abfc5c11b764ec13ea54fbef8b4421 \
+ --hash=sha256:f0cdaecd4c953bfae0b6bb64910aaaca5a424ad9c72d85cb88417bb9814f7550 \
+ --hash=sha256:f1ce721c8a7dfec21fcbdfe04e8f68174183cf4e8188e0645e92aa23985c57ff \
+ --hash=sha256:f50498891691e0864dc3da965f340fada0771f6142a378083dc4608f4ea513e2 \
+ --hash=sha256:f5ea69428fa1b49573eef0cc44a1d43bebd45ad0c611eb7d7eac760c7ae771bc \
+ --hash=sha256:f61aa92e4aad0be58eb6eb4e0c21acf32cf8065f4b2cae5665da756c4ceef982 \
+ --hash=sha256:f6e4333fb15c83f7d1482a76d45a0818897b3d33f00efd215528ff7c51b8e35d \
+ --hash=sha256:f820f24b09e3e779fe84c3c456cb4108a7aa639b0d1f02c28046e11bfcd088ed \
+ --hash=sha256:f98059e4fcd3e3e4e2d632b7cf81c2faae96c43c60b569e9c621468082f1d104 \
+ --hash=sha256:fcce033e4021347d80ed9c66dcf1e7b1546319834b74445f561d2e2221de5659
idna==3.11 \
--hash=sha256:771a87f49d9defaf64091e6e6fe9c18d4833f140bd19464795bc32d966ca37ea \
--hash=sha256:795dafcc9c04ed0c1fb032c2aa73654d8e8c5023a7df64a53f39190ada629902
inotify==0.2.12 \
--hash=sha256:9aee407f92c7d51a2ce50f3b78291a9094e334e34bd68e82bf60020795fa2c94 \
--hash=sha256:e4f1c8ec7ba5ec2a1a7fce48c0c917234af9d756495ebae7ffa00e41a305ab90
-jinja2==3.1.6 \
- --hash=sha256:0137fb05990d35f1275a587e9aee6d56da821fc83491a0fb838183be43f66d6d \
- --hash=sha256:85ece4451f492d0c13c5dd7c13a64681a86afae63a5f347908daf103ce6d2f67
-kiwisolver==1.4.9 \
- --hash=sha256:0749fd8f4218ad2e851e11cc4dc05c7cbc0cbc4267bdfdb31782e65aace4ee9c \
- --hash=sha256:0763515d4df10edf6d06a3c19734e2566368980d21ebec439f33f9eb936c07b7 \
- --hash=sha256:0856e241c2d3df4efef7c04a1e46b1936b6120c9bcf36dd216e3acd84bc4fb21 \
- --hash=sha256:0a590506f303f512dff6b7f75fd2fd18e16943efee932008fe7140e5fa91d80e \
- --hash=sha256:0ab74e19f6a2b027ea4f845a78827969af45ce790e6cb3e1ebab71bdf9f215ff \
- --hash=sha256:0ae37737256ba2de764ddc12aed4956460277f00c4996d51a197e72f62f5eec7 \
- --hash=sha256:0e4e2bf29574a6a7b7f6cb5fa69293b9f96c928949ac4a53ba3f525dffb87f9c \
- --hash=sha256:15163165efc2f627eb9687ea5f3a28137217d217ac4024893d753f46bce9de26 \
- --hash=sha256:17680d737d5335b552994a2008fab4c851bcd7de33094a82067ef3a576ff02fa \
- --hash=sha256:1a12cf6398e8a0a001a059747a1cbf24705e18fe413bc22de7b3d15c67cffe3f \
- --hash=sha256:1b11d6a633e4ed84fc0ddafd4ebfd8ea49b3f25082c04ad12b8315c11d504dc1 \
- --hash=sha256:1fa333e8b2ce4d9660f2cda9c0e1b6bafcfb2457a9d259faa82289e73ec24891 \
- --hash=sha256:2327a4a30d3ee07d2fbe2e7933e8a37c591663b96ce42a00bc67461a87d7df77 \
- --hash=sha256:2405a7d98604b87f3fc28b1716783534b1b4b8510d8142adca34ee0bc3c87543 \
- --hash=sha256:2489e4e5d7ef9a1c300a5e0196e43d9c739f066ef23270607d45aba368b91f2d \
- --hash=sha256:24c175051354f4a28c5d6a31c93906dc653e2bf234e8a4bbfb964892078898ce \
- --hash=sha256:2635d352d67458b66fd0667c14cb1d4145e9560d503219034a18a87e971ce4f3 \
- --hash=sha256:2c1a4f57df73965f3f14df20b80ee29e6a7930a57d2d9e8491a25f676e197c60 \
- --hash=sha256:2c93f00dcba2eea70af2be5f11a830a742fe6b579a1d4e00f47760ef13be247a \
- --hash=sha256:39a219e1c81ae3b103643d2aedb90f1ef22650deb266ff12a19e7773f3e5f089 \
- --hash=sha256:3b3115b2581ea35bb6d1f24a4c90af37e5d9b49dcff267eeed14c3893c5b86ab \
- --hash=sha256:40092754720b174e6ccf9e845d0d8c7d8e12c3d71e7fc35f55f3813e96376f78 \
- --hash=sha256:412f287c55a6f54b0650bd9b6dce5aceddb95864a1a90c87af16979d37c89771 \
- --hash=sha256:464415881e4801295659462c49461a24fb107c140de781d55518c4b80cb6790f \
- --hash=sha256:497d05f29a1300d14e02e6441cf0f5ee81c1ff5a304b0d9fb77423974684e08b \
- --hash=sha256:4a2899935e724dd1074cb568ce7ac0dce28b2cd6ab539c8e001a8578eb106d14 \
- --hash=sha256:4a48a2ce79d65d363597ef7b567ce3d14d68783d2b2263d98db3d9477805ba32 \
- --hash=sha256:4d1d9e582ad4d63062d34077a9a1e9f3c34088a2ec5135b1f7190c07cf366527 \
- --hash=sha256:52a15b0f35dad39862d376df10c5230155243a2c1a436e39eb55623ccbd68185 \
- --hash=sha256:540c7c72324d864406a009d72f5d6856f49693db95d1fbb46cf86febef873634 \
- --hash=sha256:5656aa670507437af0207645273ccdfee4f14bacd7f7c67a4306d0dcaeaf6eed \
- --hash=sha256:5a0f2724dfd4e3b3ac5a82436a8e6fd16baa7d507117e4279b660fe8ca38a3a1 \
- --hash=sha256:60c439763a969a6af93b4881db0eed8fadf93ee98e18cbc35bc8da868d0c4f0c \
- --hash=sha256:61874cdb0a36016354853593cffc38e56fc9ca5aa97d2c05d3dcf6922cd55a11 \
- --hash=sha256:67bb8b474b4181770f926f7b7d2f8c0248cbcb78b660fdd41a47054b28d2a752 \
- --hash=sha256:720e05574713db64c356e86732c0f3c5252818d05f9df320f0ad8380641acea5 \
- --hash=sha256:72d0eb9fba308b8311685c2268cf7d0a0639a6cd027d8128659f72bdd8a024b4 \
- --hash=sha256:767c23ad1c58c9e827b649a9ab7809fd5fd9db266a9cf02b0e926ddc2c680d58 \
- --hash=sha256:77937e5e2a38a7b48eef0585114fe7930346993a88060d0bf886086d2aa49ef5 \
- --hash=sha256:7a08b491ec91b1d5053ac177afe5290adacf1f0f6307d771ccac5de30592d198 \
- --hash=sha256:7b4da0d01ac866a57dd61ac258c5607b4cd677f63abaec7b148354d2b2cdd536 \
- --hash=sha256:7cf974dd4e35fa315563ac99d6287a1024e4dc2077b8a7d7cd3d2fb65d283134 \
- --hash=sha256:84fd60810829c27ae375114cd379da1fa65e6918e1da405f356a775d49a62bcf \
- --hash=sha256:858e4c22fb075920b96a291928cb7dea5644e94c0ee4fcd5af7e865655e4ccf2 \
- --hash=sha256:85b5352f94e490c028926ea567fc569c52ec79ce131dadb968d3853e809518c2 \
- --hash=sha256:85bd218b5ecfbee8c8a82e121802dcb519a86044c9c3b2e4aef02fa05c6da370 \
- --hash=sha256:8a1f570ce4d62d718dce3f179ee78dac3b545ac16c0c04bb363b7607a949c0d1 \
- --hash=sha256:8fdca1def57a2e88ef339de1737a1449d6dbf5fab184c54a1fca01d541317154 \
- --hash=sha256:90f47e70293fc3688b71271100a1a5453aa9944a81d27ff779c108372cf5567b \
- --hash=sha256:92a2f997387a1b79a75e7803aa7ded2cfbe2823852ccf1ba3bcf613b62ae3197 \
- --hash=sha256:9928fe1eb816d11ae170885a74d074f57af3a0d65777ca47e9aeb854a1fba386 \
- --hash=sha256:9af39d6551f97d31a4deebeac6f45b156f9755ddc59c07b402c148f5dbb6482a \
- --hash=sha256:9cf554f21be770f5111a1690d42313e140355e687e05cf82cb23d0a721a64a48 \
- --hash=sha256:a30fd6fdef1430fd9e1ba7b3398b5ee4e2887783917a687d86ba69985fb08748 \
- --hash=sha256:a31d512c812daea6d8b3be3b2bfcbeb091dbb09177706569bcfc6240dcf8b41c \
- --hash=sha256:a5d0432ccf1c7ab14f9949eec60c5d1f924f17c037e9f8b33352fa05799359b8 \
- --hash=sha256:a60ea74330b91bd22a29638940d115df9dc00af5035a9a2a6ad9399ffb4ceca5 \
- --hash=sha256:ac5a486ac389dddcc5bef4f365b6ae3ffff2c433324fb38dd35e3fab7c957999 \
- --hash=sha256:aedff62918805fb62d43a4aa2ecd4482c380dc76cd31bd7c8878588a61bd0369 \
- --hash=sha256:b34e51affded8faee0dfdb705416153819d8ea9250bbbf7ea1b249bdeb5f1122 \
- --hash=sha256:b4b4d74bda2b8ebf4da5bd42af11d02d04428b2c32846e4c2c93219df8a7987b \
- --hash=sha256:b67e6efbf68e077dd71d1a6b37e43e1a99d0bff1a3d51867d45ee8908b931098 \
- --hash=sha256:b78efa4c6e804ecdf727e580dbb9cba85624d2e1c6b5cb059c66290063bd99a9 \
- --hash=sha256:bb4ae2b57fc1d8cbd1cf7b1d9913803681ffa903e7488012be5b76dedf49297f \
- --hash=sha256:bdd1a81a1860476eb41ac4bc1e07b3f07259e6d55bbf739b79c8aaedcf512799 \
- --hash=sha256:bdee92c56a71d2b24c33a7d4c2856bd6419d017e08caa7802d2963870e315028 \
- --hash=sha256:be6a04e6c79819c9a8c2373317d19a96048e5a3f90bec587787e86a1153883c2 \
- --hash=sha256:bfc08add558155345129c7803b3671cf195e6a56e7a12f3dde7c57d9b417f525 \
- --hash=sha256:c3b22c26c6fd6811b0ae8363b95ca8ce4ea3c202d3d0975b2914310ceb1bcc4d \
- --hash=sha256:c9e7cdf45d594ee04d5be1b24dd9d49f3d1590959b2271fb30b5ca2b262c00fb \
- --hash=sha256:cb27e7b78d716c591e88e0a09a2139c6577865d7f2e152488c2cc6257f460872 \
- --hash=sha256:cc9617b46837c6468197b5945e196ee9ca43057bb7d9d1ae688101e4e1dddf64 \
- --hash=sha256:ccd09f20ccdbbd341b21a67ab50a119b64a403b09288c27481575105283c1586 \
- --hash=sha256:ce6a3a4e106cf35c2d9c4fa17c05ce0b180db622736845d4315519397a77beaf \
- --hash=sha256:d0005b053977e7b43388ddec89fa567f43d4f6d5c2c0affe57de5ebf290dc552 \
- --hash=sha256:d4188e73af84ca82468f09cadc5ac4db578109e52acb4518d8154698d3a87ca2 \
- --hash=sha256:d4efec7bcf21671db6a3294ff301d2fc861c31faa3c8740d1a94689234d1b415 \
- --hash=sha256:d75aa530ccfaa593da12834b86a0724f58bff12706659baa9227c2ccaa06264c \
- --hash=sha256:d84cd4061ae292d8ac367b2c3fa3aad11cb8625a95d135fe93f286f914f3f5a6 \
- --hash=sha256:d8aacd3d4b33b772542b2e01beb50187536967b514b00003bdda7589722d2a64 \
- --hash=sha256:d8fc5c867c22b828001b6a38d2eaeb88160bf5783c6cb4a5e440efc981ce286d \
- --hash=sha256:d976bbb382b202f71c67f77b0ac11244021cfa3f7dfd9e562eefcea2df711548 \
- --hash=sha256:dba5ee5d3981160c28d5490f0d1b7ed730c22470ff7f6cc26cfcfaacb9896a07 \
- --hash=sha256:dc1ae486f9abcef254b5618dfb4113dd49f94c68e3e027d03cf0143f3f772b61 \
- --hash=sha256:dd0a578400839256df88c16abddf9ba14813ec5f21362e1fe65022e00c883d4d \
- --hash=sha256:deed0c7258ceb4c44ad5ec7d9918f9f14fd05b2be86378d86cf50e63d1e7b771 \
- --hash=sha256:e09c2279a4d01f099f52d5c4b3d9e208e91edcbd1a175c9662a8b16e000fece9 \
- --hash=sha256:e2ea9f7ab7fbf18fffb1b5434ce7c69a07582f7acc7717720f1d69f3e806f90c \
- --hash=sha256:e6b93f13371d341afee3be9f7c5964e3fe61d5fa30f6a30eb49856935dfe4fc3 \
- --hash=sha256:eb14a5da6dc7642b0f3a18f13654847cd8b7a2550e2645a5bda677862b03ba16 \
- --hash=sha256:ed0fecd28cc62c54b262e3736f8bb2512d8dcfdc2bcf08be5f47f96bf405b145 \
- --hash=sha256:ede8c6d533bc6601a47ad4046080d36b8fc99f81e6f1c17b0ac3c2dc91ac7611 \
- --hash=sha256:efb3a45b35622bb6c16dbfab491a8f5a391fe0e9d45ef32f4df85658232ca0e2 \
- --hash=sha256:f117e1a089d9411663a3207ba874f31be9ac8eaa5b533787024dc07aeb74f464 \
- --hash=sha256:f2ba92255faa7309d06fe44c3a4a97efe1c8d640c2a79a5ef728b685762a6fd2 \
- --hash=sha256:f6008a4919fdbc0b0097089f67a1eb55d950ed7e90ce2cc3e640abadd2757a04 \
- --hash=sha256:f68208a520c3d86ea51acf688a3e3002615a7f0238002cccc17affecc86a8a54 \
- --hash=sha256:f68e4f3eeca8fb22cc3d731f9715a13b652795ef657a13df1ad0c7dc0e9731df \
- --hash=sha256:fb3b8132019ea572f4611d770991000d7f58127560c4889729248eb5852a102f \
- --hash=sha256:fb940820c63a9590d31d88b815e7a3aa5915cad3ce735ab45f0c730b39547de1 \
- --hash=sha256:fc1795ac5cd0510207482c3d1d3ed781143383b8cfd36f5c645f3897ce066220
-markupsafe==3.0.3 \
- --hash=sha256:0303439a41979d9e74d18ff5e2dd8c43ed6c6001fd40e5bf2e43f7bd9bbc523f \
- --hash=sha256:068f375c472b3e7acbe2d5318dea141359e6900156b5b2ba06a30b169086b91a \
- --hash=sha256:0bf2a864d67e76e5c9a34dc26ec616a66b9888e25e7b9460e1c76d3293bd9dbf \
- --hash=sha256:0db14f5dafddbb6d9208827849fad01f1a2609380add406671a26386cdf15a19 \
- --hash=sha256:0eb9ff8191e8498cca014656ae6b8d61f39da5f95b488805da4bb029cccbfbaf \
- --hash=sha256:0f4b68347f8c5eab4a13419215bdfd7f8c9b19f2b25520968adfad23eb0ce60c \
- --hash=sha256:1085e7fbddd3be5f89cc898938f42c0b3c711fdcb37d75221de2666af647c175 \
- --hash=sha256:116bb52f642a37c115f517494ea5feb03889e04df47eeff5b130b1808ce7c219 \
- --hash=sha256:12c63dfb4a98206f045aa9563db46507995f7ef6d83b2f68eda65c307c6829eb \
- --hash=sha256:133a43e73a802c5562be9bbcd03d090aa5a1fe899db609c29e8c8d815c5f6de6 \
- --hash=sha256:1353ef0c1b138e1907ae78e2f6c63ff67501122006b0f9abad68fda5f4ffc6ab \
- --hash=sha256:15d939a21d546304880945ca1ecb8a039db6b4dc49b2c5a400387cdae6a62e26 \
- --hash=sha256:177b5253b2834fe3678cb4a5f0059808258584c559193998be2601324fdeafb1 \
- --hash=sha256:1872df69a4de6aead3491198eaf13810b565bdbeec3ae2dc8780f14458ec73ce \
- --hash=sha256:1b4b79e8ebf6b55351f0d91fe80f893b4743f104bff22e90697db1590e47a218 \
- --hash=sha256:1b52b4fb9df4eb9ae465f8d0c228a00624de2334f216f178a995ccdcf82c4634 \
- --hash=sha256:1ba88449deb3de88bd40044603fafffb7bc2b055d626a330323a9ed736661695 \
- --hash=sha256:1cc7ea17a6824959616c525620e387f6dd30fec8cb44f649e31712db02123dad \
- --hash=sha256:218551f6df4868a8d527e3062d0fb968682fe92054e89978594c28e642c43a73 \
- --hash=sha256:26a5784ded40c9e318cfc2bdb30fe164bdb8665ded9cd64d500a34fb42067b1c \
- --hash=sha256:2713baf880df847f2bece4230d4d094280f4e67b1e813eec43b4c0e144a34ffe \
- --hash=sha256:2a15a08b17dd94c53a1da0438822d70ebcd13f8c3a95abe3a9ef9f11a94830aa \
- --hash=sha256:2f981d352f04553a7171b8e44369f2af4055f888dfb147d55e42d29e29e74559 \
- --hash=sha256:32001d6a8fc98c8cb5c947787c5d08b0a50663d139f1305bac5885d98d9b40fa \
- --hash=sha256:3524b778fe5cfb3452a09d31e7b5adefeea8c5be1d43c4f810ba09f2ceb29d37 \
- --hash=sha256:3537e01efc9d4dccdf77221fb1cb3b8e1a38d5428920e0657ce299b20324d758 \
- --hash=sha256:35add3b638a5d900e807944a078b51922212fb3dedb01633a8defc4b01a3c85f \
- --hash=sha256:38664109c14ffc9e7437e86b4dceb442b0096dfe3541d7864d9cbe1da4cf36c8 \
- --hash=sha256:3a7e8ae81ae39e62a41ec302f972ba6ae23a5c5396c8e60113e9066ef893da0d \
- --hash=sha256:3b562dd9e9ea93f13d53989d23a7e775fdfd1066c33494ff43f5418bc8c58a5c \
- --hash=sha256:457a69a9577064c05a97c41f4e65148652db078a3a509039e64d3467b9e7ef97 \
- --hash=sha256:4bd4cd07944443f5a265608cc6aab442e4f74dff8088b0dfc8238647b8f6ae9a \
- --hash=sha256:4e885a3d1efa2eadc93c894a21770e4bc67899e3543680313b09f139e149ab19 \
- --hash=sha256:4faffd047e07c38848ce017e8725090413cd80cbc23d86e55c587bf979e579c9 \
- --hash=sha256:509fa21c6deb7a7a273d629cf5ec029bc209d1a51178615ddf718f5918992ab9 \
- --hash=sha256:5678211cb9333a6468fb8d8be0305520aa073f50d17f089b5b4b477ea6e67fdc \
- --hash=sha256:591ae9f2a647529ca990bc681daebdd52c8791ff06c2bfa05b65163e28102ef2 \
- --hash=sha256:5a7d5dc5140555cf21a6fefbdbf8723f06fcd2f63ef108f2854de715e4422cb4 \
- --hash=sha256:69c0b73548bc525c8cb9a251cddf1931d1db4d2258e9599c28c07ef3580ef354 \
- --hash=sha256:6b5420a1d9450023228968e7e6a9ce57f65d148ab56d2313fcd589eee96a7a50 \
- --hash=sha256:722695808f4b6457b320fdc131280796bdceb04ab50fe1795cd540799ebe1698 \
- --hash=sha256:729586769a26dbceff69f7a7dbbf59ab6572b99d94576a5592625d5b411576b9 \
- --hash=sha256:77f0643abe7495da77fb436f50f8dab76dbc6e5fd25d39589a0f1fe6548bfa2b \
- --hash=sha256:795e7751525cae078558e679d646ae45574b47ed6e7771863fcc079a6171a0fc \
- --hash=sha256:7be7b61bb172e1ed687f1754f8e7484f1c8019780f6f6b0786e76bb01c2ae115 \
- --hash=sha256:7c3fb7d25180895632e5d3148dbdc29ea38ccb7fd210aa27acbd1201a1902c6e \
- --hash=sha256:7e68f88e5b8799aa49c85cd116c932a1ac15caaa3f5db09087854d218359e485 \
- --hash=sha256:83891d0e9fb81a825d9a6d61e3f07550ca70a076484292a70fde82c4b807286f \
- --hash=sha256:8485f406a96febb5140bfeca44a73e3ce5116b2501ac54fe953e488fb1d03b12 \
- --hash=sha256:8709b08f4a89aa7586de0aadc8da56180242ee0ada3999749b183aa23df95025 \
- --hash=sha256:8f71bc33915be5186016f675cd83a1e08523649b0e33efdb898db577ef5bb009 \
- --hash=sha256:915c04ba3851909ce68ccc2b8e2cd691618c4dc4c4232fb7982bca3f41fd8c3d \
- --hash=sha256:949b8d66bc381ee8b007cd945914c721d9aba8e27f71959d750a46f7c282b20b \
- --hash=sha256:94c6f0bb423f739146aec64595853541634bde58b2135f27f61c1ffd1cd4d16a \
- --hash=sha256:9a1abfdc021a164803f4d485104931fb8f8c1efd55bc6b748d2f5774e78b62c5 \
- --hash=sha256:9b79b7a16f7fedff2495d684f2b59b0457c3b493778c9eed31111be64d58279f \
- --hash=sha256:a320721ab5a1aba0a233739394eb907f8c8da5c98c9181d1161e77a0c8e36f2d \
- --hash=sha256:a4afe79fb3de0b7097d81da19090f4df4f8d3a2b3adaa8764138aac2e44f3af1 \
- --hash=sha256:ad2cf8aa28b8c020ab2fc8287b0f823d0a7d8630784c31e9ee5edea20f406287 \
- --hash=sha256:b8512a91625c9b3da6f127803b166b629725e68af71f8184ae7e7d54686a56d6 \
- --hash=sha256:bc51efed119bc9cfdf792cdeaa4d67e8f6fcccab66ed4bfdd6bde3e59bfcbb2f \
- --hash=sha256:bdc919ead48f234740ad807933cdf545180bfbe9342c2bb451556db2ed958581 \
- --hash=sha256:bdd37121970bfd8be76c5fb069c7751683bdf373db1ed6c010162b2a130248ed \
- --hash=sha256:be8813b57049a7dc738189df53d69395eba14fb99345e0a5994914a3864c8a4b \
- --hash=sha256:c0c0b3ade1c0b13b936d7970b1d37a57acde9199dc2aecc4c336773e1d86049c \
- --hash=sha256:c47a551199eb8eb2121d4f0f15ae0f923d31350ab9280078d1e5f12b249e0026 \
- --hash=sha256:c4ffb7ebf07cfe8931028e3e4c85f0357459a3f9f9490886198848f4fa002ec8 \
- --hash=sha256:ccfcd093f13f0f0b7fdd0f198b90053bf7b2f02a3927a30e63f3ccc9df56b676 \
- --hash=sha256:d2ee202e79d8ed691ceebae8e0486bd9a2cd4794cec4824e1c99b6f5009502f6 \
- --hash=sha256:d53197da72cc091b024dd97249dfc7794d6a56530370992a5e1a08983ad9230e \
- --hash=sha256:d6dd0be5b5b189d31db7cda48b91d7e0a9795f31430b7f271219ab30f1d3ac9d \
- --hash=sha256:d88b440e37a16e651bda4c7c2b930eb586fd15ca7406cb39e211fcff3bf3017d \
- --hash=sha256:de8a88e63464af587c950061a5e6a67d3632e36df62b986892331d4620a35c01 \
- --hash=sha256:df2449253ef108a379b8b5d6b43f4b1a8e81a061d6537becd5582fba5f9196d7 \
- --hash=sha256:e1c1493fb6e50ab01d20a22826e57520f1284df32f2d8601fdd90b6304601419 \
- --hash=sha256:e1cf1972137e83c5d4c136c43ced9ac51d0e124706ee1c8aa8532c1287fa8795 \
- --hash=sha256:e2103a929dfa2fcaf9bb4e7c091983a49c9ac3b19c9061b6d5427dd7d14d81a1 \
- --hash=sha256:e56b7d45a839a697b5eb268c82a71bd8c7f6c94d6fd50c3d577fa39a9f1409f5 \
- --hash=sha256:e8afc3f2ccfa24215f8cb28dcf43f0113ac3c37c2f0f0806d8c70e4228c5cf4d \
- --hash=sha256:e8fc20152abba6b83724d7ff268c249fa196d8259ff481f3b1476383f8f24e42 \
- --hash=sha256:eaa9599de571d72e2daf60164784109f19978b327a3910d3e9de8c97b5b70cfe \
- --hash=sha256:ec15a59cf5af7be74194f7ab02d0f59a62bdcf1a537677ce67a2537c9b87fcda \
- --hash=sha256:f190daf01f13c72eac4efd5c430a8de82489d9cff23c364c3ea822545032993e \
- --hash=sha256:f34c41761022dd093b4b6896d4810782ffbabe30f2d443ff5f083e0cbbb8c737 \
- --hash=sha256:f3e98bb3798ead92273dc0e5fd0f31ade220f59a266ffd8a4f6065e0a3ce0523 \
- --hash=sha256:f42d0984e947b8adf7dd6dde396e720934d12c506ce84eea8476409563607591 \
- --hash=sha256:f71a396b3bf33ecaa1626c255855702aca4d3d9fea5e051b41ac59a9c1c41edc \
- --hash=sha256:f9e130248f4462aaa8e2552d547f36ddadbeaa573879158d721bbd33dfe4743a \
- --hash=sha256:fed51ac40f757d41b7c48425901843666a6677e3e8eb0abcff09e4ba6e664f50
-matplotlib==3.10.8 \
- --hash=sha256:00270d217d6b20d14b584c521f810d60c5c78406dc289859776550df837dcda7 \
- --hash=sha256:0a33deb84c15ede243aead39f77e990469fff93ad1521163305095b77b72ce4a \
- --hash=sha256:113bb52413ea508ce954a02c10ffd0d565f9c3bc7f2eddc27dfe1731e71c7b5f \
- --hash=sha256:12d90df9183093fcd479f4172ac26b322b1248b15729cb57f42f71f24c7e37a3 \
- --hash=sha256:15d30132718972c2c074cd14638c7f4592bd98719e2308bccea40e0538bc0cb5 \
- --hash=sha256:18821ace09c763ec93aef5eeff087ee493a24051936d7b9ebcad9662f66501f9 \
- --hash=sha256:1ae029229a57cd1e8fe542485f27e7ca7b23aa9e8944ddb4985d0bc444f1eca2 \
- --hash=sha256:2299372c19d56bcd35cf05a2738308758d32b9eaed2371898d8f5bd33f084aa3 \
- --hash=sha256:238b7ce5717600615c895050239ec955d91f321c209dd110db988500558e70d6 \
- --hash=sha256:24d50994d8c5816ddc35411e50a86ab05f575e2530c02752e02538122613371f \
- --hash=sha256:25d380fe8b1dc32cf8f0b1b448470a77afb195438bafdf1d858bfb876f3edf7b \
- --hash=sha256:2c1998e92cd5999e295a731bcb2911c75f597d937341f3030cc24ef2733d78a8 \
- --hash=sha256:2cf5bd12cecf46908f286d7838b2abc6c91cda506c0445b8223a7c19a00df008 \
- --hash=sha256:32f8dce744be5569bebe789e46727946041199030db8aeb2954d26013a0eb26b \
- --hash=sha256:37b3c1cc42aa184b3f738cfa18c1c1d72fd496d85467a6cf7b807936d39aa656 \
- --hash=sha256:3a48a78d2786784cc2413e57397981fb45c79e968d99656706018d6e62e57958 \
- --hash=sha256:3ab4aabc72de4ff77b3ec33a6d78a68227bf1123465887f9905ba79184a1cc04 \
- --hash=sha256:3c624e43ed56313651bc18a47f838b60d7b8032ed348911c54906b130b20071b \
- --hash=sha256:3f2e409836d7f5ac2f1c013110a4d50b9f7edc26328c108915f9075d7d7a91b6 \
- --hash=sha256:3f5c3e4da343bba819f0234186b9004faba952cc420fbc522dc4e103c1985908 \
- --hash=sha256:41703cc95688f2516b480f7f339d8851a6035f18e100ee6a32bc0b8536a12a9c \
- --hash=sha256:495672de149445ec1b772ff2c9ede9b769e3cb4f0d0aa7fa730d7f59e2d4e1c1 \
- --hash=sha256:4cf267add95b1c88300d96ca837833d4112756045364f5c734a2276038dae27d \
- --hash=sha256:56271f3dac49a88d7fca5060f004d9d22b865f743a12a23b1e937a0be4818ee1 \
- --hash=sha256:595ba4d8fe983b88f0eec8c26a241e16d6376fe1979086232f481f8f3f67494c \
- --hash=sha256:5f62550b9a30afde8c1c3ae450e5eb547d579dd69b25c2fc7a1c67f934c1717a \
- --hash=sha256:646d95230efb9ca614a7a594d4fcacde0ac61d25e37dd51710b36477594963ce \
- --hash=sha256:64fcc24778ca0404ce0cb7b6b77ae1f4c7231cdd60e6778f999ee05cbd581b9a \
- --hash=sha256:6be43b667360fef5c754dda5d25a32e6307a03c204f3c0fc5468b78fa87b4160 \
- --hash=sha256:6da7c2ce169267d0d066adcf63758f0604aa6c3eebf67458930f9d9b79ad1db1 \
- --hash=sha256:83d282364ea9f3e52363da262ce32a09dfe241e4080dcedda3c0db059d3c1f11 \
- --hash=sha256:9153c3292705be9f9c64498a8872118540c3f4123d1a1c840172edf262c8be4a \
- --hash=sha256:99eefd13c0dc3b3c1b4d561c1169e65fe47aab7b8158754d7c084088e2329466 \
- --hash=sha256:a0a7f52498f72f13d4a25ea70f35f4cb60642b466cbb0a9be951b5bc3f45a486 \
- --hash=sha256:a2b336e2d91a3d7006864e0990c83b216fcdca64b5a6484912902cef87313d78 \
- --hash=sha256:a48f2b74020919552ea25d222d5cc6af9ca3f4eb43a93e14d068457f545c2a17 \
- --hash=sha256:ad3d9833a64cf48cc4300f2b406c3d0f4f4724a91c0bd5640678a6ba7c102077 \
- --hash=sha256:b44d07310e404ba95f8c25aa5536f154c0a8ec473303535949e52eb71d0a1565 \
- --hash=sha256:b53285e65d4fa4c86399979e956235deb900be5baa7fc1218ea67fbfaeaadd6f \
- --hash=sha256:b5a2b97dbdc7d4f353ebf343744f1d1f1cca8aa8bfddb4262fcf4306c3761d50 \
- --hash=sha256:b9a5ca4ac220a0cdd1ba6bcba3608547117d30468fefce49bb26f55c1a3d5c58 \
- --hash=sha256:bab485bcf8b1c7d2060b4fcb6fc368a9e6f4cd754c9c2fea281f4be21df394a2 \
- --hash=sha256:c108a1d6fa78a50646029cb6d49808ff0fc1330fda87fa6f6250c6b5369b6645 \
- --hash=sha256:d56a1efd5bfd61486c8bc968fa18734464556f0fb8e51690f4ac25d85cbbbbc2 \
- --hash=sha256:d9050fee89a89ed57b4fb2c1bfac9a3d0c57a0d55aed95949eedbc42070fea39 \
- --hash=sha256:dd80ecb295460a5d9d260df63c43f4afbdd832d725a531f008dad1664f458adf \
- --hash=sha256:e8ea3e2d4066083e264e75c829078f9e149fa119d27e19acd503de65e0b13149 \
- --hash=sha256:eb3823f11823deade26ce3b9f40dcb4a213da7a670013929f31d5f5ed1055b22 \
- --hash=sha256:ee40c27c795bda6a5292e9cff9890189d32f7e3a0bf04e0e3c9430c4a00c37df \
- --hash=sha256:efb30e3baaea72ce5928e32bab719ab4770099079d66726a62b11b1ef7273be4 \
- --hash=sha256:f254d118d14a7f99d616271d6c3c27922c092dac11112670b157798b89bf4933 \
- --hash=sha256:f89c151aab2e2e23cb3fe0acad1e8b82841fd265379c4cecd0f3fcb34c15e0f6 \
- --hash=sha256:f97aeb209c3d2511443f8797e3e5a569aebb040d4f8bc79aa3ee78a8fb9e3dd8 \
- --hash=sha256:f9b587c9c7274c1613a30afabf65a272114cd6cdbe67b3406f818c79d7ab2e2a \
- --hash=sha256:fb061f596dad3a0f52b60dc6a5dec4a0c300dec41e058a7efe09256188d170b7
-mpmath==1.3.0 \
- --hash=sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f \
- --hash=sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c
-networkx==3.1 \
- --hash=sha256:4f33f68cb2afcf86f28a45f43efc27a9386b535d567d2127f8f61d51dec58d36 \
- --hash=sha256:de346335408f84de0eada6ff9fafafff9bcda11f0a0dfaa931133debb146ab61
-numpy==1.26.4 \
- --hash=sha256:03a8c78d01d9781b28a6989f6fa1bb2c4f2d51201cf99d3dd875df6fbd96b23b \
- --hash=sha256:08beddf13648eb95f8d867350f6a018a4be2e5ad54c8d8caed89ebca558b2818 \
- --hash=sha256:1af303d6b2210eb850fcf03064d364652b7120803a0b872f5211f5234b399f20 \
- --hash=sha256:1dda2e7b4ec9dd512f84935c5f126c8bd8b9f2fc001e9f54af255e8c5f16b0e0 \
- --hash=sha256:2a02aba9ed12e4ac4eb3ea9421c420301a0c6460d9830d74a9df87efa4912010 \
- --hash=sha256:2e4ee3380d6de9c9ec04745830fd9e2eccb3e6cf790d39d7b98ffd19b0dd754a \
- --hash=sha256:3373d5d70a5fe74a2c1bb6d2cfd9609ecf686d47a2d7b1d37a8f3b6bf6003aea \
- --hash=sha256:47711010ad8555514b434df65f7d7b076bb8261df1ca9bb78f53d3b2db02e95c \
- --hash=sha256:4c66707fabe114439db9068ee468c26bbdf909cac0fb58686a42a24de1760c71 \
- --hash=sha256:50193e430acfc1346175fcbdaa28ffec49947a06918b7b92130744e81e640110 \
- --hash=sha256:52b8b60467cd7dd1e9ed082188b4e6bb35aa5cdd01777621a1658910745b90be \
- --hash=sha256:60dedbb91afcbfdc9bc0b1f3f402804070deed7392c23eb7a7f07fa857868e8a \
- --hash=sha256:62b8e4b1e28009ef2846b4c7852046736bab361f7aeadeb6a5b89ebec3c7055a \
- --hash=sha256:666dbfb6ec68962c033a450943ded891bed2d54e6755e35e5835d63f4f6931d5 \
- --hash=sha256:675d61ffbfa78604709862923189bad94014bef562cc35cf61d3a07bba02a7ed \
- --hash=sha256:679b0076f67ecc0138fd2ede3a8fd196dddc2ad3254069bcb9faf9a79b1cebcd \
- --hash=sha256:7349ab0fa0c429c82442a27a9673fc802ffdb7c7775fad780226cb234965e53c \
- --hash=sha256:7ab55401287bfec946ced39700c053796e7cc0e3acbef09993a9ad2adba6ca6e \
- --hash=sha256:7e50d0a0cc3189f9cb0aeb3a6a6af18c16f59f004b866cd2be1c14b36134a4a0 \
- --hash=sha256:95a7476c59002f2f6c590b9b7b998306fba6a5aa646b1e22ddfeaf8f78c3a29c \
- --hash=sha256:96ff0b2ad353d8f990b63294c8986f1ec3cb19d749234014f4e7eb0112ceba5a \
- --hash=sha256:9fad7dcb1aac3c7f0584a5a8133e3a43eeb2fe127f47e3632d43d677c66c102b \
- --hash=sha256:9ff0f4f29c51e2803569d7a51c2304de5554655a60c5d776e35b4a41413830d0 \
- --hash=sha256:a354325ee03388678242a4d7ebcd08b5c727033fcff3b2f536aea978e15ee9e6 \
- --hash=sha256:a4abb4f9001ad2858e7ac189089c42178fcce737e4169dc61321660f1a96c7d2 \
- --hash=sha256:ab47dbe5cc8210f55aa58e4805fe224dac469cde56b9f731a4c098b91917159a \
- --hash=sha256:afedb719a9dcfc7eaf2287b839d8198e06dcd4cb5d276a3df279231138e83d30 \
- --hash=sha256:b3ce300f3644fb06443ee2222c2201dd3a89ea6040541412b8fa189341847218 \
- --hash=sha256:b97fe8060236edf3662adfc2c633f56a08ae30560c56310562cb4f95500022d5 \
- --hash=sha256:bfe25acf8b437eb2a8b2d49d443800a5f18508cd811fea3181723922a8a82b07 \
- --hash=sha256:cd25bcecc4974d09257ffcd1f098ee778f7834c3ad767fe5db785be9a4aa9cb2 \
- --hash=sha256:d209d8969599b27ad20994c8e41936ee0964e6da07478d6c35016bc386b66ad4 \
- --hash=sha256:d5241e0a80d808d70546c697135da2c613f30e28251ff8307eb72ba696945764 \
- --hash=sha256:edd8b5fe47dab091176d21bb6de568acdd906d1887a4584a15a9a96a1dca06ef \
- --hash=sha256:f870204a840a60da0b12273ef34f7051e98c3b5961b61b0c2c1be6dfd64fbcd3 \
- --hash=sha256:ffa75af20b44f8dba823498024771d5ac50620e6915abac414251bd971b4529f
-opencv-python==4.11.0.86 \
- --hash=sha256:03d60ccae62304860d232272e4a4fda93c39d595780cb40b161b310244b736a4 \
- --hash=sha256:085ad9b77c18853ea66283e98affefe2de8cc4c1f43eda4c100cf9b2721142ec \
- --hash=sha256:1b92ae2c8852208817e6776ba1ea0d6b1e0a1b5431e971a2a0ddd2a8cc398202 \
- --hash=sha256:432f67c223f1dc2824f5e73cdfcd9db0efc8710647d4e813012195dc9122a52a \
- --hash=sha256:6b02611523803495003bd87362db3e1d2a0454a6a63025dc6658a9830570aa0d \
- --hash=sha256:810549cb2a4aedaa84ad9a1c92fbfdfc14090e2749cedf2c1589ad8359aa169b \
- --hash=sha256:9d05ef13d23fe97f575153558653e2d6e87103995d54e6a35db3f282fe1f9c66
-opencv-python-headless==4.11.0.86 \
- --hash=sha256:0e0a27c19dd1f40ddff94976cfe43066fbbe9dfbb2ec1907d66c19caef42a57b \
- --hash=sha256:48128188ade4a7e517237c8e1e11a9cdf5c282761473383e77beb875bb1e61ca \
- --hash=sha256:6c304df9caa7a6a5710b91709dd4786bf20a74d57672b3c31f7033cc638174ca \
- --hash=sha256:6efabcaa9df731f29e5ea9051776715b1bdd1845d7c9530065c7951d2a2899eb \
- --hash=sha256:996eb282ca4b43ec6a3972414de0e2331f5d9cda2b41091a49739c19fb843798 \
- --hash=sha256:a66c1b286a9de872c343ee7c3553b084244299714ebb50fbdcd76f07ebbe6c81 \
- --hash=sha256:f447d8acbb0b6f2808da71fddd29c1cdd448d2bc98f72d9bb78a7a898fc9621b
-openvino==2024.6.0 \
- --hash=sha256:05a436e2526bf6775b487712ab4a6decf7afe04183109bc7b129b95205ec4247 \
- --hash=sha256:151c6010f0613ace29087bf864dc5dd1db0afa25ab87379c735f85500545b167 \
- --hash=sha256:17cbdae3b069d9a13389b43a15b521a86fef75f350cfd0d992cc1780a6123788 \
- --hash=sha256:1d59ccc2df4a8c8515f30683f6cc0091c5b4e6a01a71ebfb1dc8f2ebdb5c83f4 \
- --hash=sha256:263496653200270b8a17456dc7bc67a198ed29b601b0ad7280d90783ba918d66 \
- --hash=sha256:2acfe89f61ec18d608c79147cedea6ba85d80783723ab9a9416891b74b15785e \
- --hash=sha256:444a6d3a746dd728654877993c30582faefcad2868c80ec261839a144b8925bc \
- --hash=sha256:46d3e20838c918668ec10ae124e92d56ff47a7c14da0d9d19d01d91fc9dece94 \
- --hash=sha256:4a94ef27022d56a159ea8677e5561bdd6f0d6c6237736e13c0f7cde71224ce7f \
- --hash=sha256:89c866fa87ddeccb982252e7d1c8516101a4acb95aaebac7e43f450112cb6942 \
- --hash=sha256:8d0390f4ab9b8f27814cf0feeddb0bba449ae8cd99933c2c2fae3011d33973b3 \
- --hash=sha256:9150c75e19606a2f2ab411271c79031726980fb6870b96445db1faca5ab64f5f \
- --hash=sha256:989c803d1f8e6d12ee8d2e3e8c7d56189d96688433b2238270dbfbf64f163f18 \
- --hash=sha256:ac812efa9c139387db534498a1e980103a35b9b2b5f3b2d1ae6ed43db51b4bba \
- --hash=sha256:c25d6820e960c08c0cc8d7823e9e6e395d762bc603772af14f0a576fab1628eb \
- --hash=sha256:c3477e833541df4f316240491d9df157c20a81487991748eb33e02e534009999 \
- --hash=sha256:c6c6a40a890b51007b20cf97d81783c4b43edfbcd4aabb06b6b04b7f2fb3137f \
- --hash=sha256:cb94f4ca9c1857bda2c5b510e519db20ea246958dbd76ddc38c133e2216231a5 \
- --hash=sha256:f7177d9a734a61a0708bcd706592f59fc2377d8efbb0bf9f8a33edfd2998a45d \
- --hash=sha256:ff5f871162299906fb517d5c9ded79f464a0341d8e2d55a037d4f9e38eed8652
-openvino-dev==2024.6.0 \
- --hash=sha256:b8f4a1baeea1b138bc9b75b53dc3bdad3307c2cd4d58526ad977df81635870c5
-openvino-telemetry==2025.2.0 \
- --hash=sha256:8bf8127218e51e99547bf38b8fb85a8b31c9bf96e6f3a82eb0b3b6a34155977c \
- --hash=sha256:bcb667e83a44f202ecf4cfa49281715c6d7e21499daec04ff853b7f964833599
packaging==26.0 \
--hash=sha256:00243ae351a257117b6a241061796684b084ed1c516a08c48a3f7e147a9d80b4 \
--hash=sha256:b36f1fef9334a5588b4166f8bcd26a14e521f2b55e6b9de3aaa80d3ff7a37529
-pillow==12.2.0 \
- --hash=sha256:00a2865911330191c0b818c59103b58a5e697cae67042366970a6b6f1b20b7f9 \
- --hash=sha256:01afa7cf67f74f09523699b4e88c73fb55c13346d212a59a2db1f86b0a63e8c5 \
- --hash=sha256:03e7e372d5240cc23e9f07deca4d775c0817bffc641b01e9c3af208dbd300987 \
- --hash=sha256:03f6fab9219220f041c74aeaa2939ff0062bd5c364ba9ce037197f4c6d498cd9 \
- --hash=sha256:042db20a421b9bafecc4b84a8b6e444686bd9d836c7fd24542db3e7df7baad9b \
- --hash=sha256:0538bd5e05efec03ae613fd89c4ce0368ecd2ba239cc25b9f9be7ed426b0af1f \
- --hash=sha256:0a34329707af4f73cf1782a36cd2289c0368880654a2c11f027bcee9052d35dd \
- --hash=sha256:0c838a5125cee37e68edec915651521191cef1e6aa336b855f495766e77a366e \
- --hash=sha256:144748b3af2d1b358d41286056d0003f47cb339b8c43a9ea42f5fea4d8c66b6e \
- --hash=sha256:1610dd6c61621ae1cf811bef44d77e149ce3f7b95afe66a4512f8c59f25d9ebe \
- --hash=sha256:1e1757442ed87f4912397c6d35a0db6a7b52592156014706f17658ff58bbf795 \
- --hash=sha256:22db17c68434de69d8ecfc2fe821569195c0c373b25cccb9cbdacf2c6e53c601 \
- --hash=sha256:25373b66e0dd5905ed63fa3cae13c82fbddf3079f2c8bf15c6fb6a35586324c1 \
- --hash=sha256:2bb4a8d594eacdfc59d9e5ad972aa8afdd48d584ffd5f13a937a664c3e7db0ed \
- --hash=sha256:2c727a6d53cb0018aadd8018c2b938376af27914a68a492f59dfcaca650d5eea \
- --hash=sha256:2d192a155bbcec180f8564f693e6fd9bccff5a7af9b32e2e4bf8c9c69dbad6b5 \
- --hash=sha256:2e589959f10d9824d39b350472b92f0ce3b443c0a3442ebf41c40cb8361c5b97 \
- --hash=sha256:2e5a76d03a6c6dcef67edabda7a52494afa4035021a79c8558e14af25313d453 \
- --hash=sha256:325ca0528c6788d2a6c3d40e3568639398137346c3d6e66bb61db96b96511c98 \
- --hash=sha256:34c0d99ecccea270c04882cb3b86e7b57296079c9a4aff88cb3b33563d95afaa \
- --hash=sha256:390ede346628ccc626e5730107cde16c42d3836b89662a115a921f28440e6a3b \
- --hash=sha256:394167b21da716608eac917c60aa9b969421b5dcbbe02ae7f013e7b85811c69d \
- --hash=sha256:3997232e10d2920a68d25191392e3a4487d8183039e1c74c2297f00ed1c50705 \
- --hash=sha256:3adc9215e8be0448ed6e814966ecf3d9952f0ea40eb14e89a102b87f450660d8 \
- --hash=sha256:3e080565d8d7c671db5802eedfb438e5565ffa40115216eabb8cd52d0ecce024 \
- --hash=sha256:4a6c9fa44005fa37a91ebfc95d081e8079757d2e904b27103f4f5fa6f0bf78c0 \
- --hash=sha256:4bfd07bc812fbd20395212969e41931001fd59eb55a60658b0e5710872e95286 \
- --hash=sha256:4e6c62e9d237e9b65fac06857d511e90d8461a32adcc1b9065ea0c0fa3a28150 \
- --hash=sha256:50d8520da2a6ce0af445fa6d648c4273c3eeefbc32d7ce049f22e8b5c3daecc2 \
- --hash=sha256:51c4167c34b0d8ba05b547a3bb23578d0ba17b80a5593f93bd8ecb123dd336a3 \
- --hash=sha256:56a3f9c60a13133a98ecff6197af34d7824de9b7b38c3654861a725c970c197b \
- --hash=sha256:56b25336f502b6ed02e889f4ece894a72612fe885889a6e8c4c80239ff6e5f5f \
- --hash=sha256:57850958fe9c751670e49b2cecf6294acc99e562531f4bd317fa5ddee2068463 \
- --hash=sha256:58f62cc0f00fd29e64b29f4fd923ffdb3859c9f9e6105bfc37ba1d08994e8940 \
- --hash=sha256:5c0a9f29ca8e79f09de89293f82fc9b0270bb4af1d58bc98f540cc4aedf03166 \
- --hash=sha256:5cdfebd752ec52bf5bb4e35d9c64b40826bc5b40a13df7c3cda20a2c03a0f5ed \
- --hash=sha256:5d04bfa02cc2d23b497d1e90a0f927070043f6cbf303e738300532379a4b4e0f \
- --hash=sha256:5d2fd0fa6b5d9d1de415060363433f28da8b1526c1c129020435e186794b3795 \
- --hash=sha256:62f5409336adb0663b7caa0da5c7d9e7bdbaae9ce761d34669420c2a801b2780 \
- --hash=sha256:632ff19b2778e43162304d50da0181ce24ac5bb8180122cbe1bf4673428328c7 \
- --hash=sha256:6562ace0d3fb5f20ed7290f1f929cae41b25ae29528f2af1722966a0a02e2aa1 \
- --hash=sha256:673aa32138f3e7531ccdbca7b3901dba9b70940a19ccecc6a37c77d5fdeb05b5 \
- --hash=sha256:6a6e67ea2e6feda684ed370f9a1c52e7a243631c025ba42149a2cc5934dec295 \
- --hash=sha256:6a9adfc6d24b10f89588096364cc726174118c62130c817c2837c60cf08a392b \
- --hash=sha256:6bb77b2dcb06b20f9f4b4a8454caa581cd4dd0643a08bacf821216a16d9c8354 \
- --hash=sha256:6e6b2a0c538fc200b38ff9eb6628228b77908c319a005815f2dde585a0664b60 \
- --hash=sha256:71cde9a1e1551df7d34a25462fc60325e8a11a82cc2e2f54578e5e9a1e153d65 \
- --hash=sha256:7371b48c4fa448d20d2714c9a1f775a81155050d383333e0a6c15b1123dda005 \
- --hash=sha256:766cef22385fa1091258ad7e6216792b156dc16d8d3fa607e7545b2b72061f1c \
- --hash=sha256:7b14cc0106cd9aecda615dd6903840a058b4700fcb817687d0ee4fc8b6e389be \
- --hash=sha256:7f84204dee22a783350679a0333981df803dac21a0190d706a50475e361c93f5 \
- --hash=sha256:8023abc91fba39036dbce14a7d6535632f99c0b857807cbbbf21ecc9f4717f06 \
- --hash=sha256:80b2da48193b2f33ed0c32c38140f9d3186583ce7d516526d462645fd98660ae \
- --hash=sha256:8297651f5b5679c19968abefd6bb84d95fe30ef712eb1b2d9b2d31ca61267f4c \
- --hash=sha256:88d387ff40b3ff7c274947ed3125dedf5262ec6919d83946753b5f3d7c67ea4c \
- --hash=sha256:88ddbc66737e277852913bd1e07c150cc7bb124539f94c4e2df5344494e0a612 \
- --hash=sha256:8bd7903a5f2a4545f6fd5935c90058b89d30045568985a71c79f5fd6edf9b91e \
- --hash=sha256:8be29e59487a79f173507c30ddf57e733a357f67881430449bb32614075a40ab \
- --hash=sha256:8c984051042858021a54926eb597d6ee3012393ce9c181814115df4c60b9a808 \
- --hash=sha256:8cbeb542b2ebc6fcdacabf8aca8c1a97c9b3ad3927d46b8723f9d4f033288a0f \
- --hash=sha256:8e9c4f5b3c546fa3458a29ab22646c1c6c787ea8f5ef51300e5a60300736905e \
- --hash=sha256:90e6f81de50ad6b534cab6e5aef77ff6e37722b2f5d908686f4a5c9eba17a909 \
- --hash=sha256:975385f4776fafde056abb318f612ef6285b10a1f12b8570f3647ad0d74b48ec \
- --hash=sha256:9a8a34cc89c67a65ea7437ce257cea81a9dad65b29805f3ecee8c8fe8ff25ffe \
- --hash=sha256:9aba9a17b623ef750a4d11b742cbafffeb48a869821252b30ee21b5e91392c50 \
- --hash=sha256:9f08483a632889536b8139663db60f6724bfcb443c96f1b18855860d7d5c0fd4 \
- --hash=sha256:a4e8f36e677d3336f35089648c8955c51c6d386a13cf6ee9c189c5f5bd713a9f \
- --hash=sha256:a52edc8bfff4429aaabdf4d9ee0daadbbf8562364f940937b941f87a4290f5ff \
- --hash=sha256:a830b1a40919539d07806aa58e1b114df53ddd43213d9c8b75847eee6c0182b5 \
- --hash=sha256:aa88ccfe4e32d362816319ed727a004423aab09c5cea43c01a4b435643fa34eb \
- --hash=sha256:af73337013e0b3b46f175e79492d96845b16126ddf79c438d7ea7ff27783a414 \
- --hash=sha256:b1c1fbd8a5a1af3412a0810d060a78b5136ec0836c8a4ef9aa11807f2a22f4e1 \
- --hash=sha256:b85f66ae9eb53e860a873b858b789217ba505e5e405a24b85c0464822fe88032 \
- --hash=sha256:b86024e52a1b269467a802258c25521e6d742349d760728092e1bc2d135b4d76 \
- --hash=sha256:bd9c0c7a0c681a347b3194c500cb1e6ca9cab053ea4d82a5cf45b6b754560136 \
- --hash=sha256:bfa9c230d2fe991bed5318a5f119bd6780cda2915cca595393649fc118ab895e \
- --hash=sha256:d362d1878f00c142b7e1a16e6e5e780f02be8195123f164edf7eddd911eefe7c \
- --hash=sha256:d5d38f1411c0ed9f97bcb49b7bd59b6b7c314e0e27420e34d99d844b9ce3b6f3 \
- --hash=sha256:dac8d77255a37e81a2efcbd1fc05f1c15ee82200e6c240d7e127e25e365c39ea \
- --hash=sha256:dd025009355c926a84a612fecf58bb315a3f6814b17ead51a8e48d3823d9087f \
- --hash=sha256:deede7c263feb25dba4e82ea23058a235dcc2fe1f6021025dc71f2b618e26104 \
- --hash=sha256:e74473c875d78b8e9d5da2a70f7099549f9eb37ded4e2f6a463e60125bccd176 \
- --hash=sha256:ee3120ae9dff32f121610bb08e4313be87e03efeadfc6c0d18f89127e24d0c24 \
- --hash=sha256:eedf4b74eda2b5a4b2b2fb4c006d6295df3bf29e459e198c90ea48e130dc75c3 \
- --hash=sha256:efd8c21c98c5cc60653bcb311bef2ce0401642b7ce9d09e03a7da87c878289d4 \
- --hash=sha256:f1c943e96e85df3d3478f7b691f229887e143f81fedab9b20205349ab04d73ed \
- --hash=sha256:f278f034eb75b4e8a13a54a876cc4a5ab39173d2cdd93a638e1b467fc545ac43 \
- --hash=sha256:f3f40b3c5a968281fd507d519e444c35f0ff171237f4fdde090dd60699458421 \
- --hash=sha256:f490f9368b6fc026f021db16d7ec2fbf7d89e2edb42e8ec09d2c60505f5729c7 \
- --hash=sha256:fb043ee2f06b41473269765c2feae53fc2e2fbf96e5e22ca94fb5ad677856f06 \
- --hash=sha256:fc3d34d4a8fbec3e88a79b92e5465e0f9b842b628675850d860b8bd300b159f5
-pip==26.0.1 \
- --hash=sha256:bdb1b08f4274833d62c1aa29e20907365a2ceb950410df15fc9521bad440122b \
- --hash=sha256:c4037d8a277c89b320abe636d59f91e6d0922d08a05b60e85e53b296613346d8
-polars==1.37.1 \
- --hash=sha256:0309e2a4633e712513401964b4d95452f124ceabf7aec6db50affb9ced4a274e \
- --hash=sha256:377fed8939a2f1223c1563cfabdc7b4a3d6ff846efa1f2ddeb8644fafd9b1aff
-polars-runtime-32==1.37.1 \
- --hash=sha256:04f5d5a2f013dca7391b7d8e7672fa6d37573a87f1d45d3dd5f0d9b5565a4b0f \
- --hash=sha256:0b8d4d73ea9977d3731927740e59d814647c5198bdbe359bcf6a8bfce2e79771 \
- --hash=sha256:55f2c4847a8d2e267612f564de7b753a4bde3902eaabe7b436a0a4abf75949a0 \
- --hash=sha256:68779d4a691da20a5eb767d74165a8f80a2bdfbde4b54acf59af43f7fa028d8f \
- --hash=sha256:a8362d11ac5193b994c7e9048ffe22ccfb976699cfbf6e128ce0302e06728894 \
- --hash=sha256:c682bf83f5f352e5e02f5c16c652c48ca40442f07b236f30662b22217320ce76 \
- --hash=sha256:da3d3642ae944e18dd17109d2a3036cb94ce50e5495c5023c77b1599d4c861bc \
- --hash=sha256:fbfde7c0ca8209eeaed546e4a32cca1319189aa61c5f0f9a2b4494262bd0c689 \
- --hash=sha256:fc82b5bbe70ca1a4b764eed1419f6336752d6ba9fc1245388d7f8b12438afa2c
-protobuf==5.29.6 \
- --hash=sha256:36ade6ff88212e91aef4e687a971a11d7d24d6948a66751abc1b3238648f5d05 \
- --hash=sha256:62e8a3114992c7c647bce37dcc93647575fc52d50e48de30c6fcb28a6a291eb1 \
- --hash=sha256:6b9edb641441b2da9fa8f428760fc136a49cf97a52076010cf22a2ff73438a86 \
- --hash=sha256:76e07e6567f8baf827137e8d5b8204b6c7b6488bbbff1bf0a72b383f77999c18 \
- --hash=sha256:7e6ad413275be172f67fdee0f43484b6de5a904cc1c3ea9804cb6fe2ff366eda \
- --hash=sha256:831e2da16b6cc9d8f1654c041dd594eda43391affd3c03a91bea7f7f6da106d6 \
- --hash=sha256:a8866b2cff111f0f863c1b3b9e7572dc7eaea23a7fae27f6fc613304046483e6 \
- --hash=sha256:b5a169e664b4057183a34bdc424540e86eea47560f3c123a0d64de4e137f9269 \
- --hash=sha256:cb4c86de9cd8a7f3a256b9744220d87b847371c6b2f10bde87768918ef33ba49 \
- --hash=sha256:da9ee6a5424b6b30fd5e45c5ea663aef540ca95f9ad99d1e887e819cdf9b8723 \
- --hash=sha256:e3387f44798ac1106af0233c04fb8abf543772ff241169946f698b3a9a3d3ab9
-psutil==7.2.1 \
- --hash=sha256:05cc68dbb8c174828624062e73078e7e35406f4ca2d0866c272c2410d8ef06d1 \
- --hash=sha256:08a2f175e48a898c8eb8eace45ce01777f4785bc744c90aa2cc7f2fa5462a266 \
- --hash=sha256:0d67c1822c355aa6f7314d92018fb4268a76668a536f133599b91edd48759442 \
- --hash=sha256:2ceae842a78d1603753561132d5ad1b2f8a7979cb0c283f5b52fb4e6e14b1a79 \
- --hash=sha256:35630d5af80d5d0d49cfc4d64c1c13838baf6717a13effb35869a5919b854cdf \
- --hash=sha256:3fce5f92c22b00cdefd1645aa58ab4877a01679e901555067b1bd77039aa589f \
- --hash=sha256:494c513ccc53225ae23eec7fe6e1482f1b8a44674241b54561f755a898650679 \
- --hash=sha256:5e38404ca2bb30ed7267a46c02f06ff842e92da3bb8c5bfdadbd35a5722314d8 \
- --hash=sha256:81442dac7abfc2f4f4385ea9e12ddf5a796721c0f6133260687fec5c3780fa49 \
- --hash=sha256:923f8653416604e356073e6e0bccbe7c09990acef442def2f5640dd0faa9689f \
- --hash=sha256:93f3f7b0bb07711b49626e7940d6fe52aa9940ad86e8f7e74842e73189712129 \
- --hash=sha256:99a4cd17a5fdd1f3d014396502daa70b5ec21bf4ffe38393e152f8e449757d67 \
- --hash=sha256:ab2b98c9fc19f13f59628d94df5cc4cc4844bc572467d113a8b517d634e362c6 \
- --hash=sha256:b1b0671619343aa71c20ff9767eced0483e4fc9e1f489d50923738caf6a03c17 \
- --hash=sha256:b2e953fcfaedcfbc952b44744f22d16575d3aa78eb4f51ae74165b4e96e55f42 \
- --hash=sha256:ba9f33bb525b14c3ea563b2fd521a84d2fa214ec59e3e6a2858f78d0844dd60d \
- --hash=sha256:cfbe6b40ca48019a51827f20d830887b3107a74a79b01ceb8cc8de4ccb17b672 \
- --hash=sha256:d34d2ca888208eea2b5c68186841336a7f5e0b990edec929be909353a202768a \
- --hash=sha256:ea46c0d060491051d39f0d2cff4f98d5c72b288289f57a21556cc7d504db37fc \
- --hash=sha256:f7583aec590485b43ca601dd9cea0dcd65bd7bb21d30ef4ddbf4ea6b5ed1bdd3 \
- --hash=sha256:f78baafb38436d5a128f837fab2d92c276dfb48af01a240b861ae02b2413ada8
-pyparsing==3.3.2 \
- --hash=sha256:850ba148bd908d7e2411587e247a1e4f0327839c40e2e5e6d05a007ecc69911d \
- --hash=sha256:c777f4d763f140633dcb6d8a3eda953bf7a214dc4eff598413c070bcdc117cbc
pyproject-hooks==1.2.0 \
--hash=sha256:1e859bd5c40fae9448642dd871adf459e5e2084186e8d2c2a79a824c970da1f8 \
--hash=sha256:9e5c6bfa8dcc30091c74b0cf803c81fdd29d94f01992a7707bc97babb1141913
-python-dateutil==2.9.0.post0 \
- --hash=sha256:37dd54208da7e1cd875388217d5e00ebd4179249f90fb72437e91a35459a0ad3 \
- --hash=sha256:a8b2bc7bffae282281c8140a97d3aa9c14da0b136dfe83f850eea9a5f7470427
pyyaml==6.0.3 \
--hash=sha256:00c4bdeba853cc34e7dd471f16b4114f4162dc03e6b7afcc2128711f0eca823c \
--hash=sha256:0150219816b6a1fa26fb4699fb7daa9caf09eb1999f3b70fb6e786805e80375a \
@@ -812,123 +223,54 @@ pyyaml==6.0.3 \
requests==2.33.1 \
--hash=sha256:18817f8c57c6263968bc123d237e3b8b08ac046f5456bd1e307ee8f4250d3517 \
--hash=sha256:4e6d1ef462f3626a1f0a0a9c42dd93c63bad33f9f1c1937509b8c5c8718ab56a
-scipy==1.15.3 \
- --hash=sha256:05dc6abcd105e1a29f95eada46d4a3f251743cfd7d3ae8ddb4088047f24ea477 \
- --hash=sha256:06efcba926324df1696931a57a176c80848ccd67ce6ad020c810736bfd58eb1c \
- --hash=sha256:0a769105537aa07a69468a0eefcd121be52006db61cdd8cac8a0e68980bbb723 \
- --hash=sha256:0bdd905264c0c9cfa74a4772cdb2070171790381a5c4d312c973382fc6eaf730 \
- --hash=sha256:0ff17c0bb1cb32952c09217d8d1eed9b53d1463e5f1dd6052c7857f83127d539 \
- --hash=sha256:14ed70039d182f411ffc74789a16df3835e05dc469b898233a245cdfd7f162cb \
- --hash=sha256:185cd3d6d05ca4b44a8f1595af87f9c372bb6acf9c808e99aa3e9aa03bd98cf6 \
- --hash=sha256:18aaacb735ab38b38db42cb01f6b92a2d0d4b6aabefeb07f02849e47f8fb3594 \
- --hash=sha256:1c832e1bd78dea67d5c16f786681b28dd695a8cb1fb90af2e27580d3d0967e92 \
- --hash=sha256:263961f658ce2165bbd7b99fa5135195c3a12d9bef045345016b8b50c315cb82 \
- --hash=sha256:271e3713e645149ea5ea3e97b57fdab61ce61333f97cfae392c28ba786f9bb49 \
- --hash=sha256:2c620736bcc334782e24d173c0fdbb7590a0a436d2fdf39310a8902505008759 \
- --hash=sha256:34716e281f181a02341ddeaad584205bd2fd3c242063bd3423d61ac259ca7eba \
- --hash=sha256:39cb9c62e471b1bb3750066ecc3a3f3052b37751c7c3dfd0fd7e48900ed52982 \
- --hash=sha256:3ac07623267feb3ae308487c260ac684b32ea35fd81e12845039952f558047b8 \
- --hash=sha256:3b0334816afb8b91dab859281b1b9786934392aa3d527cd847e41bb6f45bee65 \
- --hash=sha256:40e54d5c7e7ebf1aa596c374c49fa3135f04648a0caabcb66c52884b943f02b4 \
- --hash=sha256:50f9e62461c95d933d5c5ef4a1f2ebf9a2b4e83b0db374cb3f1de104d935922e \
- --hash=sha256:52092bc0472cfd17df49ff17e70624345efece4e1a12b23783a1ac59a1b728ed \
- --hash=sha256:5380741e53df2c566f4d234b100a484b420af85deb39ea35a1cc1be84ff53a5c \
- --hash=sha256:5e721fed53187e71d0ccf382b6bf977644c533e506c4d33c3fb24de89f5c3ed5 \
- --hash=sha256:6487aa99c2a3d509a5227d9a5e889ff05830a06b2ce08ec30df6d79db5fcd5c5 \
- --hash=sha256:6ac6310fdbfb7aa6612408bd2f07295bcbd3fda00d2d702178434751fe48e019 \
- --hash=sha256:6cfd56fc1a8e53f6e89ba3a7a7251f7396412d655bca2aa5611c8ec9a6784a1e \
- --hash=sha256:6db907c7368e3092e24919b5e31c76998b0ce1684d51a90943cb0ed1b4ffd6c1 \
- --hash=sha256:721d6b4ef5dc82ca8968c25b111e307083d7ca9091bc38163fb89243e85e3889 \
- --hash=sha256:76ad1fb5f8752eabf0fa02e4cc0336b4e8f021e2d5f061ed37d6d264db35e3ca \
- --hash=sha256:79167bba085c31f38603e11a267d862957cbb3ce018d8b38f79ac043bc92d825 \
- --hash=sha256:795c46999bae845966368a3c013e0e00947932d68e235702b5c3f6ea799aa8c9 \
- --hash=sha256:7e11270a000969409d37ed399585ee530b9ef6aa99d50c019de4cb01e8e54e62 \
- --hash=sha256:8c9ed3ba2c8a2ce098163a9bdb26f891746d02136995df25227a20e71c396ebb \
- --hash=sha256:993439ce220d25e3696d1b23b233dd010169b62f6456488567e830654ee37a6b \
- --hash=sha256:9d61e97b186a57350f6d6fd72640f9e99d5a4a2b8fbf4b9ee9a841eab327dc13 \
- --hash=sha256:9db984639887e3dffb3928d118145ffe40eff2fa40cb241a306ec57c219ebbbb \
- --hash=sha256:9e2abc762b0811e09a0d3258abee2d98e0c703eee49464ce0069590846f31d40 \
- --hash=sha256:a345928c86d535060c9c2b25e71e87c39ab2f22fc96e9636bd74d1dbf9de448c \
- --hash=sha256:ad3432cb0f9ed87477a8d97f03b763fd1d57709f1bbde3c9369b1dff5503b253 \
- --hash=sha256:ae48a786a28412d744c62fd7816a4118ef97e5be0bee968ce8f0a2fba7acf3bb \
- --hash=sha256:aef683a9ae6eb00728a542b796f52a5477b78252edede72b8327a886ab63293f \
- --hash=sha256:b90ab29d0c37ec9bf55424c064312930ca5f4bde15ee8619ee44e69319aab163 \
- --hash=sha256:c05045d8b9bfd807ee1b9f38761993297b10b245f012b11b13b91ba8945f7e45 \
- --hash=sha256:c9deabd6d547aee2c9a81dee6cc96c6d7e9a9b1953f74850c179f91fdc729cb7 \
- --hash=sha256:dde4fc32993071ac0c7dd2d82569e544f0bdaff66269cb475e0f369adad13f11 \
- --hash=sha256:eae3cf522bc7df64b42cad3925c876e1b0b6c35c1337c93e12c0f366f55b0eaf \
- --hash=sha256:ed7284b21a7a0c8f1b6e5977ac05396c0d008b89e05498c8b7e8f4a1423bba0e \
- --hash=sha256:f77f853d584e72e874d87357ad70f44b437331507d1c311457bed8ed2b956126
-six==1.17.0 \
- --hash=sha256:4721f391ed90541fddacab5acf947aa0d3dc7d27b2e1e8eda2be8970586c3274 \
- --hash=sha256:ff70335d468e7eb6ec65b95b99d3a2836546063f63acc5171de367e834932a81
-sympy==1.14.0 \
- --hash=sha256:d3d3fe8df1e5a0b42f0e7bdf50541697dbe7d23746e894990c030e2b05e72517 \
- --hash=sha256:e091cc3e99d2141a0ba2847328f5479b05d94a6635cb96148ccb3f34671bd8f5
-tomli==1.1.0 \
- --hash=sha256:33d7984738f8bb699c9b0a816eb646a8178a69eaa792d258486776a5d21b8ca5 \
- --hash=sha256:f4a182048010e89cbec0ae4686b21f550a7f2903f665e34a6de58ec15424f919
-torch==2.10.0+cpu \
- --hash=sha256:0e286fcf6ce0cc7b204396c9b4ea0d375f1f0c3e752f68ce3d3aeb265511db8c \
- --hash=sha256:0fedcb1a77e8f2aaf7bfd21591bf6d1e0b207473268c9be16b17cb7783253969 \
- --hash=sha256:106dd1930cb30a4a337366ba3f9b25318ebf940f51fd46f789281dd9e736bdc4 \
- --hash=sha256:170a0623108055be5199370335cf9b41ba6875b3cb6f086db4aee583331a4899 \
- --hash=sha256:179451716487f8cb09b56459667fa1f5c4c0946c1e75fbeae77cfc40a5768d87 \
- --hash=sha256:17a09465bab2aab8f0f273410297133d8d8fb6dd84dccbd252ca4a4f3a111847 \
- --hash=sha256:1cfcb9b1558c6e52dffd0d4effce83b13c5ae5d97338164c372048c21f9cfccb \
- --hash=sha256:21cb5436978ef47c823b7a813ff0f8c2892e266cfe0f1d944879b5fba81bf4e1 \
- --hash=sha256:23882f8d882460aca809882fc42f5e343bf07585274f929ced00177d1be1eb67 \
- --hash=sha256:2adc71fe471e98a608723bfc837f7e1929885ebb912c693597711e139c1cda41 \
- --hash=sha256:31ae44836c8b9bbd1a3943d29c7c7457709ddf7c6173aa34aefe9d2203e4c405 \
- --hash=sha256:3eaa727e6a73affa61564d86b9d03191df45c8650d0666bd3d57c8597ef61e78 \
- --hash=sha256:4fcd8b4cc2ae20f2b7749fb275349c55432393868778c2d50a08e81d5ee5591e \
- --hash=sha256:5af75e5f49de21b0bdf7672bc27139bd285f9e8dbcabe2d617a2eb656514ac36 \
- --hash=sha256:6c6f0df770144907092a0d067048d96ed4f278a6c840376d2ff0e27e7579b925 \
- --hash=sha256:8d316e5bf121f1eab1147e49ad0511a9d92e4c45cc357d1ab0bee440da71a095 \
- --hash=sha256:8de5a36371b775e2d4881ed12cc7f2de400b1ad3d728aa74a281f649f87c9b8c \
- --hash=sha256:9412bd37b70f5ebd1205242c4ba4cabae35a605947f2b30806d5c9b467936db9 \
- --hash=sha256:9accc30b56cb6756d4a9d04fcb8ebc0bb68c7d55c1ed31a8657397d316d31596 \
- --hash=sha256:a280ffaea7b9c828e0c1b9b3bd502d9b6a649dc9416997b69b84544bd469f215 \
- --hash=sha256:a28fdbcfa2fbacffec81300f24dd1bed2b0ccfdbed107a823cff12bc1db070f6 \
- --hash=sha256:aada8afc068add586464b2a55adb7cc9091eec55caf5320447204741cb6a0604 \
- --hash=sha256:b67d91326e4ed9eccbd6b7d84ed7ffa43f93103aa3f0b24145f3001f3b11b714 \
- --hash=sha256:b719da5af01b59126ac13eefd6ba3dd12d002dc0e8e79b8b365e55267a8189d3 \
- --hash=sha256:b7cb1ec66cefb90fd7b676eac72cfda3b8d4e4d0cacd7a531963bc2e0a9710ab \
- --hash=sha256:ba51ef01a510baf8fff576174f702c47e1aa54389a9f1fba323bb1a5003ff0bf \
- --hash=sha256:beadc2a6a1785b09a46daad378de91ef274b8d3eea7af0bc2d017d97f115afdf \
- --hash=sha256:c35c0de592941d4944698dbfa87271ab85d3370eca3b694943a2ab307ac34b3f \
- --hash=sha256:ce5c113d1f55f8c1f5af05047a24e50d11d293e0cbbb5bf7a75c6c761edd6eaa \
- --hash=sha256:d63ee6a80982fd73fe44bb70d97d2976e010312ff6db81d7bfb9167b06dd45b9 \
- --hash=sha256:e51994492cdb76edce29da88de3672a3022f9ef0ffd90345436948d4992be2c7 \
- --hash=sha256:e71c476517c33e7db69825a9ff46c7f47a723ec4dac5b2481cff4246d1c632be \
- --hash=sha256:ea2bcc9d1fca66974a71d4bf9a502539283f35d61fcab5a799b4e120846f1e02 \
- --hash=sha256:eb1bde1ce198f05c8770017de27e001d404499cf552aaaa014569eff56ca25c0 \
- --hash=sha256:ee40b8a4b4b2cf0670c6fd4f35a7ef23871af956fecb238fbf5da15a72650b1d \
- --hash=sha256:f8294fd2fc6dd8f4435a891a0122307a043b14b21f0dac1bca63c85bfb59e586 \
- --hash=sha256:fd215f3d0f681905c5b56b0630a3d666900a37fcc3ca5b937f95275c66f9fd9c \
- --hash=sha256:ffc8da9a1341092d6a90cb5b1c1a33cd61abf0fb43f0cd88443c27fa372c26ae
-torchvision==0.25.0+cpu \
- --hash=sha256:0c2d0da9bc011a0fde1d125af396a8fbe94d99becf9d313764f24ca7657a3448 \
- --hash=sha256:2d444009c0956669ada149f61ed78f257c1cc96d259efa6acf3929ca96ceb3f0 \
- --hash=sha256:3e2ae9981e32a5b9db685659d5c7af0f04b159ff20394650a90124baf6ada51a \
- --hash=sha256:499eae1e535766391b6ee2d1e6e841239c20e2e6d88203a15b8f9f8d60a1f8bd \
- --hash=sha256:4d72a57a8f0b5146e26dac1fbfa2c905280cd04f5fcb23b9c56253506b683aeb \
- --hash=sha256:59be99d1c470ef470b134468aa6afa6f968081a503acb4ee883d70332f822e35 \
- --hash=sha256:727334e9a721cfc1ac296ce0bf9e69d9486821bfa5b1e75a8feb6f78041db481 \
- --hash=sha256:73ce04dea64914ff1110008311204c366107d651d3b04faa0dbee77efb7133b7 \
- --hash=sha256:783c8fc580bbfc159bff52f4f72cdd538e42b32956e70dffa42b940db114e151 \
- --hash=sha256:7d47d544899fabac52ebe0d4812975608fd7ab79a3d7fb6383275eb667e33f53 \
- --hash=sha256:7f851245a2687743742157988ed9c42c6e312b95bbe6cfcac9e7d0d0c28ae62f \
- --hash=sha256:813f0106eb3e268f3783da67b882458e544c6fb72f946e6ca64b5ed4e62c6a77 \
- --hash=sha256:90eec299e1f82cfaf080ccb789df3838cb9a54b57e2ebe33852cd392c692de5c \
- --hash=sha256:9212210f417888e6261c040495180f053084812cf873dedba9fc51ff4b24b2d3 \
- --hash=sha256:9511339b3b5eb75229e0b5041202e8aed9bef3b1de3a715b9fb319c9e97688fd \
- --hash=sha256:aa016ab73e06a886f72edc8929ed2ed4c85aaaa6e10500ecdef921b03129b19e \
- --hash=sha256:c1be164e93c68b2dbf460fd58975377c892dbcf3358fb72941709c3857351bba \
- --hash=sha256:c7eb5f219fdfaf1f65e68c00eb81172ab4fa08a9874dae9dad2bca360da34d0f \
- --hash=sha256:e985e12a9a232618e5a43476de5689e4b14989f5da6b93909c57afa57ec27012 \
- --hash=sha256:fb9f07f6a10f0ac24ac482ae68c6df99110b74a0d80a4c64fddc9753267d8815 \
- --hash=sha256:fe54cbd5942cd0b26a90f1748f0d4421caf67be35c281c6c3b8573733a03d630
+tomli==2.4.0 \
+ --hash=sha256:0408e3de5ec77cc7f81960c362543cbbd91ef883e3138e81b729fc3eea5b9729 \
+ --hash=sha256:0dc56fef0e2c1c470aeac5b6ca8cc7b640bb93e92d9803ddaf9ea03e198f5b0b \
+ --hash=sha256:0e0fe8a0b8312acf3a88077a0802565cb09ee34107813bba1c7cd591fa6cfc8d \
+ --hash=sha256:0f2e3955efea4d1cfbcb87bc321e00dc08d2bcb737fd1d5e398af111d86db5df \
+ --hash=sha256:133e93646ec4300d651839d382d63edff11d8978be23da4cc106f5a18b7d0576 \
+ --hash=sha256:1b168f2731796b045128c45982d3a4874057626da0e2ef1fdd722848b741361d \
+ --hash=sha256:1c8a885b370751837c029ef9bc014f27d80840e48bac415f3412e6593bbc18c1 \
+ --hash=sha256:1f776e7d669ebceb01dee46484485f43a4048746235e683bcdffacdf1fb4785a \
+ --hash=sha256:1fb2945cbe303b1419e2706e711b7113da57b7db31ee378d08712d678a34e51e \
+ --hash=sha256:20cedb4ee43278bc4f2fee6cb50daec836959aadaf948db5172e776dd3d993fc \
+ --hash=sha256:20ffd184fb1df76a66e34bd1b36b4a4641bd2b82954befa32fe8163e79f1a702 \
+ --hash=sha256:26ab906a1eb794cd4e103691daa23d95c6919cc2fa9160000ac02370cc9dd3f6 \
+ --hash=sha256:2add28aacc7425117ff6364fe9e06a183bb0251b03f986df0e78e974047571fd \
+ --hash=sha256:2b1e3b80e1d5e52e40e9b924ec43d81570f0e7d09d11081b797bc4692765a3d4 \
+ --hash=sha256:31d556d079d72db7c584c0627ff3a24c5d3fb4f730221d3444f3efb1b2514776 \
+ --hash=sha256:36b9d05b51e65b254ea6c2585b59d2c4cb91c8a3d91d0ed0f17591a29aaea54a \
+ --hash=sha256:39b0b5d1b6dd03684b3fb276407ebed7090bbec989fa55838c98560c01113b66 \
+ --hash=sha256:3cf226acb51d8f1c394c1b310e0e0e61fecdd7adcb78d01e294ac297dd2e7f87 \
+ --hash=sha256:3d895d56bd3f82ddd6faaff993c275efc2ff38e52322ea264122d72729dca2b2 \
+ --hash=sha256:413540dce94673591859c4c6f794dfeaa845e98bf35d72ed59636f869ef9f86f \
+ --hash=sha256:43e685b9b2341681907759cf3a04e14d7104b3580f808cfde1dfdb60ada85475 \
+ --hash=sha256:4cbcb367d44a1f0c2be408758b43e1ffb5308abe0ea222897d6bfc8e8281ef2f \
+ --hash=sha256:551e321c6ba03b55676970b47cb1b73f14a0a4dce6a3e1a9458fd6d921d72e95 \
+ --hash=sha256:5572e41282d5268eb09a697c89a7bee84fae66511f87533a6f88bd2f7b652da9 \
+ --hash=sha256:5aa48d7c2356055feef06a43611fc401a07337d5b006be13a30f6c58f869e3c3 \
+ --hash=sha256:5b5807f3999fb66776dbce568cc9a828544244a8eb84b84b9bafc080c99597b9 \
+ --hash=sha256:5e3f639a7a8f10069d0e15408c0b96a2a828cfdec6fca05296ebcdcc28ca7c76 \
+ --hash=sha256:685306e2cc7da35be4ee914fd34ab801a6acacb061b6a7abca922aaf9ad368da \
+ --hash=sha256:75c2f8bbddf170e8effc98f5e9084a8751f8174ea6ccf4fca5398436e0320bc8 \
+ --hash=sha256:7b438885858efd5be02a9a133caf5812b8776ee0c969fea02c45e8e3f296ba51 \
+ --hash=sha256:7d49c66a7d5e56ac959cb6fc583aff0651094ec071ba9ad43df785abc2320d86 \
+ --hash=sha256:7d6d9a4aee98fac3eab4952ad1d73aee87359452d1c086b5ceb43ed02ddb16b8 \
+ --hash=sha256:84d081fbc252d1b6a982e1870660e7330fb8f90f676f6e78b052ad4e64714bf0 \
+ --hash=sha256:8768715ffc41f0008abe25d808c20c3d990f42b6e2e58305d5da280ae7d1fa3b \
+ --hash=sha256:920b1de295e72887bafa3ad9f7a792f811847d57ea6b1215154030cf131f16b1 \
+ --hash=sha256:9a08144fa4cba33db5255f9b74f0b89888622109bd2776148f2597447f92a94e \
+ --hash=sha256:a26d7ff68dfdb9f87a016ecfd1e1c2bacbe3108f4e0f8bcd2228ef9a766c787d \
+ --hash=sha256:aa89c3f6c277dd275d8e243ad24f3b5e701491a860d5121f2cdd399fbb31fc9c \
+ --hash=sha256:b5ef256a3fd497d4973c11bf142e9ed78b150d36f5773f1ca6088c230ffc5867 \
+ --hash=sha256:b6c78bdf37764092d369722d9946cb65b8767bfa4110f902a1b2542d8d173c8a \
+ --hash=sha256:bbb1b10aa643d973366dc2cb1ad94f99c1726a02343d43cbc011edbfac579e7c \
+ --hash=sha256:c084ad935abe686bd9c898e62a02a19abfc9760b5a79bc29644463eaf2840cb0 \
+ --hash=sha256:c73add4bb52a206fd0c0723432db123c0c75c280cbd67174dd9d2db228ebb1b4 \
+ --hash=sha256:cae9c19ed12d4e8f3ebf46d1a75090e4c0dc16271c5bce1c833ac168f08fb614 \
+ --hash=sha256:d20b797a5c1ad80c516e41bc1fb0443ddb5006e9aaa7bda2d71978346aeb9132 \
+ --hash=sha256:d3d1654e11d724760cdb37a3d7691f0be9db5fbdaef59c9f532aabf87006dbaa \
+ --hash=sha256:d878f2a6707cc9d53a1be1414bbb419e629c3d6e67f69230217bb663e76b5087
tornado==6.5.5 \
--hash=sha256:192b8f3ea91bd7f1f50c06955416ed76c6b72f96779b962f07f911b91e8d30e9 \
--hash=sha256:2c9a876e094109333f888539ddb2de4361743e5d21eece20688e3e351e4990a6 \
@@ -940,21 +282,11 @@ tornado==6.5.5 \
--hash=sha256:65a7f1d46d4bb41df1ac99f5fcb685fb25c7e61613742d5108b010975a9a6521 \
--hash=sha256:dd3eafaaeec1c7f2f8fdcd5f964e8907ad788fe8a5a32c4426fbbdda621223b7 \
--hash=sha256:e74c92e8e65086b338fd56333fb9a68b9f6f2fe7ad532645a290a464bcf46be5
-typing-extensions==4.15.0 \
- --hash=sha256:0cea48d173cc12fa28ecabc3b837ea3cf6f38c6d1136f85cbaaf598984861466 \
- --hash=sha256:f0fa19c6845758ab08074a0cfa8b7aecb71c999ca73d62883bc25cc018c4e548
-ultralytics==8.4.34 \
- --hash=sha256:e932da39976879ddf4f398993bd44587c1f5aafaac138efab38d34030680c113 \
- --hash=sha256:fe94bf945c2c69b4df245ee9327dc3c7ce6d06ec9c647843331b0d497a48db78
-ultralytics-thop==2.0.18 \
- --hash=sha256:21103bcd39cc9928477dc3d9374561749b66a1781b35f46256c8d8c4ac01d9cf \
- --hash=sha256:2bb44851ad224b116c3995b02dd5e474a5ccf00acf237fe0edb9e1506ede04ec
urllib3==2.6.3 \
--hash=sha256:1b62b6884944a57dbe321509ab94fd4d3b307075e0c2eae991ac71ee15ad38ed \
--hash=sha256:bf272323e553dfb2e87d9bfd225ca7b0f467b919d7bbd355436d3fd37cb0acd4
-vdms==0.0.23 \
- --hash=sha256:2dec153f8cb33f27cdb9ab33125198195fd29a6777cf890fc735774ed9327874 \
- --hash=sha256:7cd93242df644947c009559f31b4e19eb1dab6b62ab783ce5a2a54a4ad392f57
-wheel==0.46.3 \
- --hash=sha256:4b399d56c9d9338230118d705d9737a2a468ccca63d5e813e2a4fc7815d8bc4d \
- --hash=sha256:e3e79874b07d776c40bd6033f8ddf76a7dad46a7b8aa1b2787a83083519a1803
+
+# The following packages are considered to be unsafe in a requirements file:
+pip==26.0.1 \
+ --hash=sha256:bdb1b08f4274833d62c1aa29e20907365a2ceb950410df15fc9521bad440122b \
+ --hash=sha256:c4037d8a277c89b320abe636d59f91e6d0922d08a05b60e85e53b296613346d8
diff --git a/video/resources/models/download_yolo.py b/video/resources/models/download_yolo.py
deleted file mode 100644
index 9dae0b5..0000000
--- a/video/resources/models/download_yolo.py
+++ /dev/null
@@ -1,130 +0,0 @@
-import os
-from pathlib import Path
-
-from ultralytics import YOLO
-
-
-def str2bool(in_val):
- if isinstance(in_val, bool):
- return in_val
-
- if not isinstance(in_val, str):
- raise ValueError(f"{in_val} is not a bool or string")
-
- if in_val.title() == "True":
- return True
- else:
- return False
-
-
-model_precision_object = "FP16"
-half_flag = True
-dynamic_flag = True
-CUSTOM_MODEL_FLAG = str2bool(os.getenv("CUSTOM_MODEL_FLAG", False))
-DEVICE = os.environ.get("DEVICE", "CPU")
-MODEL_NAME = os.environ.get("MODEL_NAME", "yolo11n")
-if DEVICE == "GPU":
- batch_size = int(os.environ.get("GPU_BATCH_SIZE", 1))
- run_platform_name = "engine"
- os.environ["CUDA_VISIBLE_DEVICES"] = "0"
- print("[!] USING GPU & TENSORRT")
-else:
- # batch_size = 8
- batch_size = int(os.environ.get("CPU_BATCH_SIZE", 1)) # 8
- run_platform_name = "openvino"
- print("[!] USING CPU & OPENVINO")
-
-
-def get_model(model_dir, run_platform, device_input, batch=1):
- final_model_path = f"{model_dir}/{MODEL_NAME}.pt"
- pt_detection_model = YOLO(final_model_path, verbose=False, task="detect")
- if run_platform == "openvino":
- pt_detection_model.export(
- format="openvino",
- half=half_flag,
- dynamic=dynamic_flag,
- device=device_input,
- batch=batch,
- )
-
- final_model_path = f"{model_dir}/{MODEL_NAME}_openvino_model/"
- object_detection_model = YOLO(
- final_model_path,
- verbose=False,
- task="detect",
- )
-
- # det_ov_model = core.read_model(final_model_path+"yolo11n.xml")
- # ov_config = {hints.performance_mode: hints.PerformanceMode.LATENCY}
- # if device == "GPU":
- # ov_config["GPU_DISABLE_WINOGRAD_CONVOLUTION"] = "YES"
- # compiled_model = core.compile_model(det_ov_model, device, ov_config)
- # object_detection_model.predictor.model.ov_compiled_model = compiled_model
-
- elif run_platform == "engine":
- pt_detection_model.export(
- format="engine",
- half=half_flag,
- imgsz=[7680, 4320], # Max dimensions (8K-[W,H]-[7680,4320])
- dynamic=dynamic_flag,
- device=device_input,
- simplify=True,
- batch=batch,
- )
-
- final_model_path = f"{model_dir}/{MODEL_NAME}.engine"
- object_detection_model = YOLO(
- final_model_path,
- verbose=False,
- task="detect",
- )
-
- elif run_platform == "onnx":
- from torch import cuda
- from ultralytics.utils.checks import check_requirements
-
- check_requirements(
- "onnxruntime-gpu"
- if cuda.is_available() and device_input != "cpu"
- else "onnxruntime"
- )
-
- final_model_path = f"{model_dir}/{MODEL_NAME}.onnx"
- pt_detection_model.export(
- format="onnx",
- half=half_flag,
- dynamic=dynamic_flag,
- device=device_input,
- simplify=True,
- batch=batch,
- )
-
- object_detection_model = YOLO(final_model_path, verbose=False, task="detect")
-
- elif run_platform == "pytorch":
- object_detection_model = pt_detection_model
- if DEVICE == "GPU":
- object_detection_model.to("cuda")
- else:
- object_detection_model.to(device_input)
-
- else:
- raise ValueError(f"[!] Model for {run_platform} is not implemented.")
-
- return object_detection_model, final_model_path
-
-
-if __name__ == "__main__":
- if CUSTOM_MODEL_FLAG:
- dir_path = "/home/resources/models/ultralytics/custom_models"
- else:
- dir_path = (
- f"/home/resources/models/ultralytics/{MODEL_NAME}/{model_precision_object}"
- )
-
- ydir = Path(dir_path)
-
- device_input = DEVICE.lower() if DEVICE == "CPU" else "cuda"
- _, _ = get_model(ydir, run_platform_name, device_input, batch=batch_size)
-
- os.remove(f"{ydir}/{MODEL_NAME}.pt")
diff --git a/video/segment.py b/video/segment.py
index dc5a7d8..6cedf16 100755
--- a/video/segment.py
+++ b/video/segment.py
@@ -7,7 +7,14 @@
from tornado import gen, web
from tornado.concurrent import run_on_executor
-from utils import safely_join_path
+
+
+def safely_join_path(base_dir, add_path):
+ safe_base = os.path.abspath(base_dir)
+ candidate_path = os.path.abspath(os.path.join(safe_base, add_path))
+ if not candidate_path.startswith(safe_base + os.sep):
+ raise ValueError(f"Invalid path: {candidate_path}")
+ return candidate_path
class SegmentHandler(web.RequestHandler):
diff --git a/video/source_watcher.py b/video/source_watcher.py
new file mode 100644
index 0000000..9abcac3
--- /dev/null
+++ b/video/source_watcher.py
@@ -0,0 +1,317 @@
+import logging
+import multiprocessing as mp
+import os
+import sys
+import time
+import traceback
+from multiprocessing.managers import BaseManager
+from pathlib import Path
+
+import requests
+import yaml
+from inotify.adapters import Inotify
+
+logger = logging.getLogger(__name__)
+
+BACKEND_URL = os.getenv("BACKEND_URL", "http://fastapi-service:8000")
+DEBUG = os.environ["DEBUG"]
+DEBUG_FLAG = True if DEBUG == "1" else False
+ACCEPTED_VIDEO_FORMATS = [".mp4", ".mkv", ".avi"]
+num_workers = mp.cpu_count() // 2
+empty_timeout = 5 * 60
+
+
+# --- HELPER FUNCTIONS ---
+def is_file_ready(filepath, retries=5, delay=1):
+ """
+ Checks if a file is fully written and accessible.
+ Uses exponential backoff for retries.
+ """
+ for _ in range(retries):
+ try:
+ # Try to open the file in append mode to check for locks
+ with open(filepath, "r"):
+ return True
+ except (IOError, OSError):
+ logger.warning(f"File {filepath} is locked. Retrying in {delay}s...")
+ time.sleep(delay)
+ delay *= 2 # Exponential backoff
+ return False
+
+
+def connect_to_app():
+ # is_ready = False
+ # while not is_ready:
+ with requests.Session() as session:
+ while True:
+ try:
+ res = session.get(f"{BACKEND_URL}/status")
+
+ if res.status_code == 200:
+ # Expected format: {"cam_1": {"status": "Ready"}, ...}
+ data = res.json()
+
+ is_ready = any(status == "Ready" for status in data.values())
+ if is_ready:
+ break
+ except requests.exceptions.RequestException:
+ logger.info("BACKEND: Waiting for fastapi-service ...")
+
+ time.sleep(1)
+
+
+# --- WATCHERS & MANAGERS ---
+class QueueManager(BaseManager):
+ pass
+
+
+def worker_process(queue):
+ """Consumes tasks from the queue and sends them to the backend."""
+ while True:
+ try:
+ # Wait for a task; timeout prevents hanging on shutdown
+ task = queue.get(timeout=10)
+ if task is None:
+ break # Sentinel value to stop worker
+
+ source, camera_name = task
+ start_stream_processor(source, camera_name)
+ except mp.queues.Empty:
+ continue
+ except Exception as e:
+ logger.error(f"Worker error: {e}")
+
+
+def watch_video_files(queue, watch_dir):
+ # Initial scan of files
+ with os.scandir(watch_dir) as entries:
+ for entry in entries:
+ # entry.is_file() and entry.name are retrieved in one go
+ if entry.is_file() and any(
+ entry.name.endswith(ext) for ext in ACCEPTED_VIDEO_FORMATS
+ ):
+ source = entry.path # Full path is already available
+
+ if is_file_ready(source):
+ queue.put((source, None))
+ logger.info(f"{source} added to queue: {time.time()}")
+ # start_stream_processor(source, None)
+
+ # Watch watch_dir for new files
+ i = Inotify()
+ i.add_watch(watch_dir)
+ logger.info("START ADDING VIDEO FILES TO WATCHED DIRECTORY")
+
+ for event in i.event_gen(yield_nones=False):
+ (_, type_names, path, filename) = event
+
+ # New file created in watched directory
+ # IN_CLOSE_WRITE is better than IN_CREATE because it triggers
+ # only after the writing process finishes and closes the file.
+ if "IN_CLOSE_WRITE" in type_names and any(
+ filename.endswith(ext) for ext in ACCEPTED_VIDEO_FORMATS
+ ):
+ source = os.path.join(path, filename)
+
+ if is_file_ready(source):
+ queue.put((source, None))
+ logger.info(f"{source} added to queue: {time.time()}")
+ # start_stream_processor(source, None)
+ else:
+ logger.error(f"Failed to access {source} after retries. Skipping.")
+
+
+def retrieve_camera_details(queue, config_path):
+ """
+ Watches config file and adds only new streams (via camera_name) to queue
+ """
+ unique_camera_names = set()
+
+ def read_config_and_queue():
+ try:
+ with open(config_path, "r") as inFile:
+ config = yaml.safe_load(inFile)
+
+ if config:
+ for camera_name, camera_details in config.items():
+ # run_processor(camera_details["url"], camera_name=camera_name)
+ if isinstance(camera_details, dict) and "url" in camera_details:
+ source = camera_details["url"]
+ if source not in unique_camera_names:
+ logger.info(f"{source} added to queue: {time.time()}")
+ queue.put((source, camera_name))
+ unique_camera_names.add(source)
+ # start_stream_processor(source, camera_name)
+ else:
+ logger.warning(f"Invalid entry: {camera_name}")
+ except Exception:
+ e = traceback.format_exc()
+ logger.info(f"Unexpected error processing config file: {e}")
+
+ # Initial scan of config file
+ read_config_and_queue()
+
+ # Watch for changes to file
+ i = Inotify()
+ config_dir = os.path.dirname(os.path.abspath(config_path))
+ i.add_watch(config_dir)
+
+ for event in i.event_gen(yield_nones=False):
+ (_, type_names, path, filename) = event
+
+ if filename == os.path.basename(config_path) and "IN_CLOSE_WRITE" in type_names:
+ time.sleep(0.5)
+ read_config_and_queue()
+
+
+def unified_watcher(queue, watch_dir, config_path):
+ """
+ Watches directory for both videos and config changes.
+ """
+ unique_camera_names = set()
+ config_filename = os.path.basename(config_path)
+
+ def read_config():
+ try:
+ with open(config_path, "r") as f:
+ config = yaml.safe_load(f)
+ if config:
+ for name, details in config.items():
+ url = details.get("url")
+ if url and url not in unique_camera_names:
+ logger.info(f"CAMERA: {name} added to queue: {time.time()}")
+ queue.put((url, name))
+ unique_camera_names.add(url)
+ except Exception as e:
+ logger.error(f"CONFIG ERROR: {e}")
+
+ # Initial scan of config file
+ read_config()
+
+ # Initial scan of video files
+ with os.scandir(watch_dir) as entries:
+ for entry in entries:
+ if entry.is_file() and any(
+ entry.name.endswith(ext) for ext in ACCEPTED_VIDEO_FORMATS
+ ):
+ if is_file_ready(entry.path):
+ queue.put((entry.path, None))
+
+ # Unified Event Loop
+ i = Inotify()
+ i.add_watch(watch_dir)
+ logger.info(f"UNIFIED WATCHER: Monitoring {watch_dir} for videos and config...")
+
+ for event in i.event_gen(yield_nones=False):
+ (_, type_names, path, filename) = event
+ full_path = os.path.join(path, filename)
+
+ # Handle Config Update (Supports Atomic Saves/Moves)
+ if filename == config_filename:
+ if any(ev in type_names for ev in ["IN_CLOSE_WRITE", "IN_MOVED_TO"]):
+ time.sleep(0.2)
+ read_config()
+
+ # Handle New Video Files
+ elif any(filename.endswith(ext) for ext in ACCEPTED_VIDEO_FORMATS):
+ if "IN_CLOSE_WRITE" in type_names:
+ if is_file_ready(full_path):
+ queue.put((full_path, None))
+ logger.info(f"VIDEO: {full_path} added to queue: {time.time()}")
+
+
+# --- STREAMER ---
+def start_stream_processor(source, camera_name):
+ if camera_name is None:
+ camera_name = Path(source).stem
+
+ payload = {"url": str(source), "name": camera_name}
+
+ for attempt in range(1, 11):
+ try:
+ # FastAPI stream_processor
+ res = requests.post(f"{BACKEND_URL}/stream", json=payload)
+
+ # Raise an exception for 4xx or 5xx status codes
+ res.raise_for_status()
+
+ if res.status_code == 200:
+ logger.info(f"Started {source} process on attempt {attempt}.")
+ return res
+
+ except requests.exceptions.HTTPError as e:
+ # Handle specific API errors (e.g., 400 Bad Request, 500 Internal Server Error)
+ logger.info(f"HTTP Error: {e.response.status_code} - {e.response.text}")
+
+ # If it's a client error (4xx), retrying likely won't help
+ if 400 <= e.response.status_code < 500:
+ break
+
+ except requests.exceptions.RequestException as e:
+ # Handle connection issues, timeouts, and DNS errors
+ logger.info(f"Connection attempt {attempt} failed: {e}")
+
+ except Exception:
+ e = traceback.format_exc()
+ logger.info(f"Unexpected error processing {source}: {e}")
+
+ time.sleep(1)
+
+
+# --- MAIN FUNCTION ---
+def main(watch_folder=os.getcwd()):
+ # Wait until App is ready
+ connect_to_app()
+
+ if DEBUG_FLAG:
+ logger.info("[TIMING],start_watchandsend,," + str(time.time()))
+
+ # Setup the Shared Queue and Manager
+ file_queue = mp.Queue()
+ QueueManager.register("get_file_queue", callable=lambda: file_queue)
+
+ # Start the manager on port 5005
+ manager = QueueManager(address=("0.0.0.0", 5005), authkey=b"password123")
+ manager.start()
+ logger.info("Queue Server started at 0.0.0.0:5005")
+
+ # Define worker processes
+ processes = [
+ mp.Process( # Process that retrieves camera info from config file (added to file_queue)
+ target=unified_watcher,
+ args=(file_queue, watch_folder, "/home/camera_config.yaml"),
+ name="UnifiedWatcher",
+ daemon=True,
+ ),
+ ]
+
+ for _ in range(num_workers):
+ p = mp.Process(target=worker_process, args=(file_queue,), daemon=True)
+ processes.append(p)
+
+ # 4. Start processed and Keep the server alive
+ try:
+ # Start all processes
+ for p in processes:
+ p.start()
+ while True:
+ time.sleep(1)
+ except KeyboardInterrupt:
+ logger.info("Stopping services ...")
+ finally:
+ # Cleanup
+ for p in processes:
+ if p.is_alive():
+ p.terminate()
+ p.join() # Ensure fully closed
+
+ manager.shutdown()
+ if DEBUG_FLAG:
+ logger.info("[TIMING],end_watchandsend,," + str(time.time()))
+
+
+if __name__ == "__main__":
+ if len(sys.argv) == 2:
+ main(sys.argv[1])
+ else:
+ raise ValueError("Invalid input. Please provide watch directory.")
diff --git a/video/thumbnail.py b/video/thumbnail.py
index 9a0d236..342e054 100755
--- a/video/thumbnail.py
+++ b/video/thumbnail.py
@@ -7,7 +7,14 @@
from tornado import gen, web
from tornado.concurrent import run_on_executor
-from utils import safely_join_path
+
+
+def safely_join_path(base_dir, add_path):
+ safe_base = os.path.abspath(base_dir)
+ candidate_path = os.path.abspath(os.path.join(safe_base, add_path))
+ if not candidate_path.startswith(safe_base + os.sep):
+ raise ValueError(f"Invalid path: {candidate_path}")
+ return candidate_path
class ThumbnailHandler(web.RequestHandler):
diff --git a/video/watch_and_send2vdms.py b/video/watch_and_send2vdms.py
deleted file mode 100644
index 2bdc3b8..0000000
--- a/video/watch_and_send2vdms.py
+++ /dev/null
@@ -1,115 +0,0 @@
-import multiprocessing as mp
-import os
-import queue
-import subprocess
-import sys
-import time
-
-import yaml
-from inotify.adapters import Inotify
-
-num_workers = mp.cpu_count() // 2
-
-DEBUG = os.environ["DEBUG"]
-DEBUG_FLAG = True if DEBUG == "1" else False
-
-
-def run_processor(path_or_url, camera_name=None):
- cmd = [
- sys.executable,
- "/home/process_stream.py",
- path_or_url,
- # camera_name,
- ]
- if camera_name is not None:
- cmd.append(camera_name)
-
- subprocess.run(cmd, check=True)
-
-
-def watch_video_files(queue, watch_dir):
- # Get files already in watch_dir
- for filename in os.listdir(watch_dir):
- if any(filename.endswith(ext) for ext in [".mp4", ".mkv", ".avi"]):
- source = os.path.join(watch_dir, filename)
- print(f"{source} added to queue: {time.time()}", flush=True)
- queue.put((source, None))
-
- # Watch watch_dir for new files
- i = Inotify()
- i.add_watch(watch_dir)
- print("START ADDING VIDEO FILES TO WATCHED DIRECTORY", flush=True)
-
- for event in i.event_gen(yield_nones=False):
- (_, type_names, path, filename) = event
-
- # New file created in watched directory
- # if "IN_CREATE" in type_names:
- if "IN_CLOSE_WRITE" in type_names and any(
- filename.endswith(ext) for ext in [".mp4", ".mkv", ".avi"]
- ):
- source = os.path.join(path, filename)
- print(f"{source} added to queue: {time.time()}", flush=True)
- queue.put((source, None))
-
-
-def retrieve_camera_details(queue, config_path):
- config = None
- with open(config_path, "r") as inFile:
- config = yaml.safe_load(inFile)
-
- if config is not None:
- for camera_name, camera_details in config.items():
- # run_processor(camera_details["url"], camera_name=camera_name)
- source = camera_details["url"]
- print(f"{source} added to queue: {time.time()}", flush=True)
- queue.put((source, camera_name))
-
-
-def main(watch_folder=os.getcwd()):
- if DEBUG_FLAG:
- print("[TIMING],start_watchandsend,," + str(time.time()), flush=True)
-
- file_queue = mp.Queue()
-
- # Create a process that monitors new files in watch_folder (added to file_queue)
- watcher_process = mp.Process(
- target=watch_video_files, args=(file_queue, watch_folder)
- )
-
- # Create a process that retrieves camera info from config file (added to file_queue)
- watcher_process_camera = mp.Process(
- target=retrieve_camera_details, args=(file_queue, "/home/camera_config.yaml")
- )
-
- watcher_process.start()
- watcher_process_camera.start()
-
- # Pool of workers to process video clips
- with mp.Pool(processes=num_workers) as pool:
- while True:
- try:
- path_or_url, camera_name = file_queue.get(timeout=0.5)
- # pool.apply(run_processor, (path_or_url, camera_name,))
- pool.apply_async(
- run_processor,
- (
- path_or_url,
- camera_name,
- ),
- )
- except queue.Empty:
- pass
-
- watcher_process.join()
- watcher_process_camera.join()
-
- if DEBUG_FLAG:
- print("[TIMING],end_watchandsend,," + str(time.time()), flush=True)
-
-
-if __name__ == "__main__":
- if len(sys.argv) == 2:
- main(sys.argv[1])
- else:
- raise ValueError("Invalid input. Please provide watch directory.")
+
+ {% endfor %}
+ Camera: {{ id }}
+ +
+ 
+
+