Skip to content

Remove SIMULATE_CHOOSER_COLUMNS memory hacks and expose global constants in utility expressions #1075

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
asiripanich wants to merge 1 commit into from
Closed

Remove SIMULATE_CHOOSER_COLUMNS memory hacks and expose global constants in utility expressions #1075

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 5 additions & 32 deletions activitysim/abm/models/location_choice.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -161,6 +161,7 @@ def _location_sample(
"reindex": reindex,
"land_use": state.get_dataframe("land_use"),
}
locals_d.update(state.get_global_constants())
locals_d.update(model_settings.CONSTANTS or {})

# preprocess choosers table
Expand Down Expand Up @@ -232,14 +233,7 @@ def location_sample(
chunk_tag,
trace_label,
):
# FIXME - MEMORY HACK - only include columns actually used in spec
chooser_columns = model_settings.SIMULATE_CHOOSER_COLUMNS
# Drop this when PR #1017 is merged
if ("household_id" not in chooser_columns) and (
"household_id" in persons_merged.columns
):
chooser_columns = chooser_columns + ["household_id"]
choosers = persons_merged[chooser_columns]
choosers = persons_merged

# create wrapper with keys for this lookup - in this case there is a home_zone_id in the choosers
# and a zone_id in the alternatives which get merged during interaction
Expand Down Expand Up @@ -390,17 +384,7 @@ def location_presample(
HOME_TAZ in persons_merged
) # 'TAZ' should already be in persons_merged from land_use

# FIXME - MEMORY HACK - only include columns actually used in spec
# FIXME we don't actually require that land_use provide a TAZ crosswalk
# FIXME maybe we should add it for multi-zone (from maz_taz) if missing?
chooser_columns = model_settings.SIMULATE_CHOOSER_COLUMNS
chooser_columns = [HOME_TAZ if c == HOME_MAZ else c for c in chooser_columns]
# Drop this when PR #1017 is merged
if ("household_id" not in chooser_columns) and (
"household_id" in persons_merged.columns
):
chooser_columns = chooser_columns + ["household_id"]
choosers = persons_merged[chooser_columns]
choosers = persons_merged

# create wrapper with keys for this lookup - in this case there is a HOME_TAZ in the choosers
# and a DEST_TAZ in the alternatives which get merged during interaction
Expand Down Expand Up @@ -565,11 +549,6 @@ def run_location_logsums(
mandatory=False,
)

# FIXME - MEMORY HACK - only include columns actually used in spec
persons_merged_df = logsum.filter_chooser_columns(
persons_merged_df, logsum_settings, model_settings
)

logger.info(f"Running {trace_label} with {len(location_sample_df.index)} rows")

choosers = location_sample_df.join(persons_merged_df, how="left")
Expand Down Expand Up @@ -628,14 +607,7 @@ def run_location_simulate(
"""
assert not persons_merged.empty

# FIXME - MEMORY HACK - only include columns actually used in spec
chooser_columns = model_settings.SIMULATE_CHOOSER_COLUMNS
# Drop this when PR #1017 is merged
if ("household_id" not in chooser_columns) and (
"household_id" in persons_merged.columns
):
chooser_columns = chooser_columns + ["household_id"]
choosers = persons_merged[chooser_columns]
choosers = persons_merged

alt_dest_col_name = model_settings.ALT_DEST_COL_NAME

Expand Down Expand Up @@ -666,6 +638,7 @@ def run_location_simulate(
"reindex": reindex,
"land_use": state.get_dataframe("land_use"),
}
locals_d.update(state.get_global_constants())
locals_d.update(model_settings.CONSTANTS or {})

# preprocess choosers table
Expand Down
34 changes: 22 additions & 12 deletions activitysim/abm/models/school_escorting.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@
from __future__ import annotations

import logging
import warnings
from typing import Any, Literal

import numpy as np
Expand All @@ -17,6 +18,8 @@
tracing,
workflow,
)
from pydantic import field_validator

from activitysim.core.configuration.base import PreprocessorSettings
from activitysim.core.configuration.logit import BaseLogitComponentSettings
from activitysim.core.interaction_simulate import interaction_simulate
Expand Down Expand Up @@ -335,7 +338,25 @@ class SchoolEscortSettings(BaseLogitComponentSettings, extra="forbid"):
GENDER_WEIGHT: float = 10.0
AGE_WEIGHT: float = 1.0

SIMULATE_CHOOSER_COLUMNS: list[str] | None = None
SIMULATE_CHOOSER_COLUMNS: Any | None = None
"""Was used to help reduce the memory needed for the model.
Setting is now obsolete and doesn't do anything.
Functionality was replaced by util.drop_unused_columns

.. deprecated:: 1.4
"""

@field_validator("SIMULATE_CHOOSER_COLUMNS", mode="before")
@classmethod
def _warn_simulate_chooser_columns(cls, v):
if v is not None:
warnings.warn(
"SIMULATE_CHOOSER_COLUMNS is deprecated and replaced by "
"util.drop_unused_columns; value will be ignored.",
DeprecationWarning,
stacklevel=2,
)
return None

SPEC: None = None
"""The school escort model does not use this setting."""
Expand Down Expand Up @@ -465,17 +486,6 @@ def school_escorting(
# else:
# locals_dict.pop("_sharrow_skip", None)

# reduce memory by limiting columns if selected columns are supplied
chooser_columns = model_settings.SIMULATE_CHOOSER_COLUMNS
if chooser_columns is not None:
# Drop this when PR #1017 is merged
if ("household_id" not in chooser_columns) and (
"household_id" in choosers.columns
):
chooser_columns = chooser_columns + ["household_id"]
chooser_columns = chooser_columns + participant_columns
choosers = choosers[chooser_columns]

# add previous data to stage
if stage_num >= 1:
choosers = add_prev_choices_to_choosers(
Expand Down
36 changes: 0 additions & 36 deletions activitysim/abm/models/util/logsums.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,6 @@
import pandas as pd

from activitysim.core import config, expressions, los, simulate, tracing, workflow
from activitysim.core.configuration import PydanticBase
from activitysim.core.configuration.logit import (
TourLocationComponentSettings,
TourModeComponentSettings,
Expand All @@ -16,41 +15,6 @@
logger = logging.getLogger(__name__)


def filter_chooser_columns(
choosers, logsum_settings: dict | PydanticBase, model_settings: dict | PydanticBase
):
try:
chooser_columns = logsum_settings.LOGSUM_CHOOSER_COLUMNS
except AttributeError:
chooser_columns = logsum_settings.get("LOGSUM_CHOOSER_COLUMNS", [])

if (
isinstance(model_settings, dict)
and "CHOOSER_ORIG_COL_NAME" in model_settings
and model_settings["CHOOSER_ORIG_COL_NAME"] not in chooser_columns
):
chooser_columns.append(model_settings["CHOOSER_ORIG_COL_NAME"])
if (
isinstance(model_settings, PydanticBase)
and hasattr(model_settings, "CHOOSER_ORIG_COL_NAME")
and model_settings.CHOOSER_ORIG_COL_NAME
and model_settings.CHOOSER_ORIG_COL_NAME not in chooser_columns
):
chooser_columns.append(model_settings.CHOOSER_ORIG_COL_NAME)

missing_columns = [c for c in chooser_columns if c not in choosers]
if missing_columns:
logger.debug(
"logsum.filter_chooser_columns missing_columns %s" % missing_columns
)

# ignore any columns not appearing in choosers df
chooser_columns = [c for c in chooser_columns if c in choosers]

choosers = choosers[chooser_columns]
return choosers


def compute_location_choice_logsums(
state: workflow.State,
choosers: pd.DataFrame,
Expand Down
35 changes: 2 additions & 33 deletions activitysim/abm/models/util/tour_destination.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -113,6 +113,7 @@ def _destination_sample(
"dest_col_name": skims.dest_key, # added for sharrow flows
"timeframe": "timeless",
}
locals_d.update(state.get_global_constants())
constants = model_settings.CONSTANTS
if constants is not None:
locals_d.update(constants)
Expand Down Expand Up @@ -619,23 +620,9 @@ def run_destination_sample(
chunk_size,
trace_label,
):
# FIXME - MEMORY HACK - only include columns actually used in spec (omit them pre-merge)
chooser_columns = model_settings.SIMULATE_CHOOSER_COLUMNS

# if special person id is passed
chooser_id_column = model_settings.CHOOSER_ID_COLUMN

# Drop this when PR #1017 is merged
if ("household_id" not in chooser_columns) and (
"household_id" in persons_merged.columns
):
chooser_columns = chooser_columns + ["household_id"]
persons_merged = persons_merged[
[c for c in persons_merged.columns if c in chooser_columns]
]
tours = tours[
[c for c in tours.columns if c in chooser_columns or c == chooser_id_column]
]
choosers = pd.merge(
tours, persons_merged, left_on=chooser_id_column, right_index=True, how="left"
)
Expand Down Expand Up @@ -731,11 +718,6 @@ def run_destination_logsums(

chunk_tag = "tour_destination.logsums"

# FIXME - MEMORY HACK - only include columns actually used in spec
persons_merged = logsum.filter_chooser_columns(
persons_merged, logsum_settings, model_settings
)

# merge persons into tours
choosers = pd.merge(
destination_sample,
Expand Down Expand Up @@ -798,23 +780,9 @@ def run_destination_simulate(
coefficients_file_name=model_settings.COEFFICIENTS,
)

# FIXME - MEMORY HACK - only include columns actually used in spec (omit them pre-merge)
chooser_columns = model_settings.SIMULATE_CHOOSER_COLUMNS

# if special person id is passed
chooser_id_column = model_settings.CHOOSER_ID_COLUMN

# Drop this when PR #1017 is merged
if ("household_id" not in chooser_columns) and (
"household_id" in persons_merged.columns
):
chooser_columns = chooser_columns + ["household_id"]
persons_merged = persons_merged[
[c for c in persons_merged.columns if c in chooser_columns]
]
tours = tours[
[c for c in tours.columns if c in chooser_columns or c == chooser_id_column]
]
choosers = pd.merge(
tours, persons_merged, left_on=chooser_id_column, right_index=True, how="left"
)
Expand Down Expand Up @@ -856,6 +824,7 @@ def run_destination_simulate(
"dest_col_name": skims.dest_key, # added for sharrow flows
"timeframe": "timeless",
}
locals_d.update(state.get_global_constants())
if constants is not None:
locals_d.update(constants)

Expand Down
20 changes: 2 additions & 18 deletions activitysim/abm/models/util/tour_od.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -178,6 +178,7 @@ def _od_sample(
"orig_col_name": ORIG_TAZ,
"dest_col_name": DEST_TAZ,
}
locals_d.update(state.get_global_constants())
constants = model_settings.CONSTANTS
if constants is not None:
locals_d.update(constants)
Expand Down Expand Up @@ -690,12 +691,6 @@ def run_od_sample(
)

choosers = tours
# FIXME - MEMORY HACK - only include columns actually used in spec
chooser_columns = model_settings.SIMULATE_CHOOSER_COLUMNS
# Drop this when PR #1017 is merged
if ("household_id" not in chooser_columns) and ("household_id" in choosers.columns):
chooser_columns = chooser_columns + ["household_id"]
choosers = choosers[chooser_columns]

# interaction_sample requires that choosers.index.is_monotonic_increasing
if not choosers.index.is_monotonic_increasing:
Expand Down Expand Up @@ -772,11 +767,6 @@ def run_od_logsums(
dest_id_col = model_settings.DEST_COL_NAME
tour_od_id_col = get_od_id_col(origin_id_col, dest_id_col)

# FIXME - MEMORY HACK - only include columns actually used in spec
tours_merged_df = logsum.filter_chooser_columns(
tours_merged_df, logsum_settings, model_settings
)

# merge ods into choosers table
choosers = od_sample.join(tours_merged_df, how="left")
choosers[tour_od_id_col] = (
Expand Down Expand Up @@ -952,13 +942,6 @@ def run_od_simulate(
# merge persons into tours
choosers = tours

# FIXME - MEMORY HACK - only include columns actually used in spec
chooser_columns = model_settings.SIMULATE_CHOOSER_COLUMNS
# Drop this when PR #1017 is merged
if ("household_id" not in chooser_columns) and ("household_id" in choosers.columns):
chooser_columns = chooser_columns + ["household_id"]
choosers = choosers[chooser_columns]

# interaction_sample requires that choosers.index.is_monotonic_increasing
if not choosers.index.is_monotonic_increasing:
logger.debug(
Expand Down Expand Up @@ -1009,6 +992,7 @@ def run_od_simulate(
"orig_col_name": origin_col_name,
"dest_col_name": dest_col_name,
}
locals_d.update(state.get_global_constants())
if constants is not None:
locals_d.update(constants)

Expand Down
24 changes: 0 additions & 24 deletions activitysim/abm/models/util/tour_scheduling.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,30 +24,6 @@ def run_tour_scheduling(
trace_label: str,
):

if model_settings.LOGSUM_SETTINGS:
logsum_settings = TourModeComponentSettings.read_settings_file(
state.filesystem,
str(model_settings.LOGSUM_SETTINGS),
mandatory=False,
)
logsum_columns = logsum_settings.LOGSUM_CHOOSER_COLUMNS
else:
logsum_columns = []

# - filter chooser columns for both logsums and simulate
model_columns = model_settings.SIMULATE_CHOOSER_COLUMNS
chooser_columns = logsum_columns + [
c for c in model_columns if c not in logsum_columns
]

# Drop this when PR #1017 is merged
if ("household_id" not in chooser_columns) and (
"household_id" in persons_merged.columns
):
chooser_columns = chooser_columns + ["household_id"]

persons_merged = expressions.filter_chooser_columns(persons_merged, chooser_columns)

timetable = state.get_injectable("timetable")

# - run preprocessor to annotate choosers
Expand Down
Loading
Loading