Enables array tests for asarray (#20)

* Tests asarray

* improvments

Author: xadupre

_unittests/onnx-numpy-skips.txt

@@ -1,6 +1,6 @@
 # API failures
+# see https://github.com/data-apis/array-api-tests/blob/master/numpy-skips.txt
 array_api_tests/test_creation_functions.py::test_arange
-array_api_tests/test_creation_functions.py::test_asarray_scalars
 array_api_tests/test_creation_functions.py::test_asarray_arrays
 array_api_tests/test_creation_functions.py::test_empty
 array_api_tests/test_creation_functions.py::test_empty_like

_unittests/test_array_api.sh

@@ -1,2 +1,4 @@
 export ARRAY_API_TESTS_MODULE=onnx_array_api.array_api.onnx_numpy
-pytest ../array-api-tests/array_api_tests/test_creation_functions.py::test_zeros
+# pytest ../array-api-tests/array_api_tests/test_creation_functions.py::test_asarray_arrays || exit 1
+# pytest ../array-api-tests/array_api_tests/test_creation_functions.py --help
+pytest ../array-api-tests/array_api_tests/test_creation_functions.py --hypothesis-explain --skips-file=_unittests/onnx-numpy-skips.txt || exit 1

_unittests/ut_array_api/test_onnx_numpy.py

@@ -6,7 +6,11 @@
 
 
 class TestOnnxNumpy(ExtTestCase):
-    def test_abs(self):
+    def test_empty(self):
+        c = EagerTensor(np.array([4, 5], dtype=np.int64))
+        self.assertRaise(lambda: xp.empty(c, dtype=xp.int64), RuntimeError)
+
+    def test_zeros(self):
         c = EagerTensor(np.array([4, 5], dtype=np.int64))
         mat = xp.zeros(c, dtype=xp.int64)
         matnp = mat.numpy()

_unittests/ut_npx/test_npx.py

@@ -2501,7 +2501,7 @@ def test_numpy_all(self):
         got = ref.run(None, {"A": data})
         self.assertEqualArray(y, got[0])
 
-    def test_numpy_all_empty(self):
+    def test_numpy_all_zeros(self):
         data = np.zeros((0,), dtype=np.bool_)
         y = np.all(data)
 
@@ -2513,7 +2513,7 @@ def test_numpy_all_empty(self):
         self.assertEqualArray(y, got[0])
 
     @unittest.skipIf(True, reason="ReduceMin does not support shape[axis] == 0")
-    def test_numpy_all_empty_axis_0(self):
+    def test_numpy_all_zeros_axis_0(self):
         data = np.zeros((0, 1), dtype=np.bool_)
         y = np.all(data, axis=0)
 
@@ -2535,7 +2535,13 @@ def test_numpy_all_empty_axis_1(self):
         got = ref.run(None, {"A": data})
         self.assertEqualArray(y, got[0])
 
+    @unittest.skipIf(True, reason="Fails to follow Array API")
+    def test_get_item(self):
+        a = EagerNumpyTensor(np.array([True], dtype=np.bool_))
+        i = a[0]
+        self.assertEqualArray(i.numpy(), a.numpy()[0])
+
 
 if __name__ == "__main__":
-    # TestNpx().test_numpy_all_empty_axis_0()
+    # TestNpx().test_get_item()
     unittest.main(verbosity=2)

azure-pipelines.yml

@@ -131,12 +131,12 @@ jobs:
   - script: |
       export ARRAY_API_TESTS_MODULE=onnx_array_api.array_api.onnx_numpy
       cd array-api-tests
-      python -m pytest -x array_api_tests/test_creation_functions.py --skips-file=../_unittests/onnx-numpy-skips.txt -v
+      python -m pytest -x array_api_tests/test_creation_functions.py --skips-file=../_unittests/onnx-numpy-skips.txt --hypothesis-explain
     displayName: "numpy test_creation_functions.py"
   - script: |
       export ARRAY_API_TESTS_MODULE=onnx_array_api.array_api.onnx_ort
       cd array-api-tests
-      python -m pytest -x array_api_tests/test_creation_functions.py --skips-file=../_unittests/onnx-ort-skips.txt -v
+      python -m pytest -x array_api_tests/test_creation_functions.py --skips-file=../_unittests/onnx-ort-skips.txt --hypothesis-explain
     displayName: "ort test_creation_functions.py"
   #- script: |
   #    export ARRAY_API_TESTS_MODULE=onnx_array_api.array_api.onnx_numpy

onnx_array_api/array_api/_onnx_common.py

@@ -34,7 +34,15 @@ def template_asarray(
         return a.astype(dtype=dtype)
 
     if isinstance(a, int):
-        v = TEagerTensor(np.array(a, dtype=np.int64))
+        if a is False:
+            v = TEagerTensor(np.array(False, dtype=np.bool_))
+        elif a is True:
+            v = TEagerTensor(np.array(True, dtype=np.bool_))
+        else:
+            try:
+                v = TEagerTensor(np.asarray(a, dtype=np.int64))
+            except OverflowError:
+                v = TEagerTensor(np.asarray(a, dtype=np.uint64))
     elif isinstance(a, float):
         v = TEagerTensor(np.array(a, dtype=np.float32))
     elif isinstance(a, bool):

onnx_array_api/array_api/onnx_numpy.py

@@ -29,6 +29,7 @@
     "all",
     "asarray",
     "astype",
+    "empty",
     "equal",
     "isdtype",
     "isfinite",
@@ -73,6 +74,17 @@ def ones(
     return generic_ones(shape, dtype=dtype, order=order)
 
 
+def empty(
+    shape: TensorType[ElemType.int64, "I", (None,)],
+    dtype: OptParType[DType] = DType(TensorProto.FLOAT),
+    order: OptParType[str] = "C",
+) -> TensorType[ElemType.numerics, "T"]:
+    raise RuntimeError(
+        "ONNX assumes there is no inplace implementation. "
+        "empty function is only used in that case."
+    )
+
+
 def zeros(
     shape: TensorType[ElemType.int64, "I", (None,)],
     dtype: OptParType[DType] = DType(TensorProto.FLOAT),

onnx_array_api/npx/npx_numpy_tensors.py

@@ -1,6 +1,6 @@
 from typing import Any, Callable, List, Optional, Tuple
 import numpy as np
-from onnx import ModelProto
+from onnx import ModelProto, TensorProto
 from onnx.reference import ReferenceEvaluator
 from .._helpers import np_dtype_to_tensor_dtype
 from .npx_numpy_tensors_ops import ConstantOfShape
@@ -183,6 +183,60 @@ def __array_namespace__(self, api_version: Optional[str] = None):
             f"Unable to return an implementation for api_version={api_version!r}."
         )
 
+    def __bool__(self):
+        "Implicit conversion to bool."
+        if self.dtype != DType(TensorProto.BOOL):
+            raise TypeError(
+                f"Conversion to bool only works for bool scalar, not for {self!r}."
+            )
+        if self.shape == (0,):
+            return False
+        if len(self.shape) != 0:
+            raise ValueError(
+                f"Conversion to bool only works for scalar, not for {self!r}."
+            )
+        return bool(self._tensor)
+
+    def __int__(self):
+        "Implicit conversion to bool."
+        if len(self.shape) != 0:
+            raise ValueError(
+                f"Conversion to bool only works for scalar, not for {self!r}."
+            )
+        if self.dtype not in {
+            DType(TensorProto.INT64),
+            DType(TensorProto.INT32),
+            DType(TensorProto.INT16),
+            DType(TensorProto.INT8),
+            DType(TensorProto.UINT64),
+            DType(TensorProto.UINT32),
+            DType(TensorProto.UINT16),
+            DType(TensorProto.UINT8),
+        }:
+            raise TypeError(
+                f"Conversion to int only works for int scalar, "
+                f"not for dtype={self.dtype}."
+            )
+        return int(self._tensor)
+
+    def __float__(self):
+        "Implicit conversion to bool."
+        if len(self.shape) != 0:
+            raise ValueError(
+                f"Conversion to bool only works for scalar, not for {self!r}."
+            )
+        if self.dtype not in {
+            DType(TensorProto.FLOAT),
+            DType(TensorProto.DOUBLE),
+            DType(TensorProto.FLOAT16),
+            DType(TensorProto.BFLOAT16),
+        }:
+            raise TypeError(
+                f"Conversion to int only works for float scalar, "
+                f"not for dtype={self.dtype}."
+            )
+        return float(self._tensor)
+
 
 class JitNumpyTensor(NumpyTensor, JitTensor):
     """

onnx_array_api/npx/npx_var.py

@@ -962,6 +962,7 @@ def __getitem__(self, index: Any) -> "Var":
 
         if isinstance(index, Var):
             # scenario 2
+            # TODO: fix this when index is an integer
             new_shape = cst(np.array([-1], dtype=np.int64))
             new_self = self.reshape(new_shape)
             new_index = index.reshape(new_shape)
@@ -973,6 +974,9 @@ def __getitem__(self, index: Any) -> "Var":
 
         if not isinstance(index, tuple):
             index = (index,)
+        elif len(index) == 0:
+            # The array contains a scalar and it needs to be returned.
+            return var(self, op="Identity")
 
         # only one integer?
         ni = None