Fixes asarray for the Array API (#25)

* Fixes asarray for the Array API

* move

Author: xadupre

_unittests/ut_array_api/test_hypothesis_array_api.py

@@ -2,6 +2,7 @@
 import warnings
 from os import getenv
 from functools import reduce
+import numpy as np
 from operator import mul
 from hypothesis import given
 from onnx_array_api.ext_test_case import ExtTestCase
@@ -89,24 +90,49 @@ def test_scalar_strategies(self):
 
         args_np = []
 
+        xx = self.xps.arrays(dtype=dtypes["integer_dtypes"], shape=shapes(self.xps))
+        kws = array_api_kwargs(dtype=strategies.none() | self.xps.scalar_dtypes())
+
         @given(
-            x=self.xps.arrays(dtype=dtypes["integer_dtypes"], shape=shapes(self.xps)),
-            kw=array_api_kwargs(dtype=strategies.none() | self.xps.scalar_dtypes()),
+            x=xx,
+            kw=kws,
         )
-        def fct(x, kw):
+        def fctnp(x, kw):
+            asa1 = np.asarray(x)
+            asa2 = np.asarray(x, **kw)
+            self.assertEqual(asa1.shape, asa2.shape)
             args_np.append((x, kw))
 
-        fct()
+        fctnp()
         self.assertEqual(len(args_np), 100)
 
         args_onxp = []
 
         xshape = shapes(self.onxps)
         xx = self.onxps.arrays(dtype=dtypes_onnx["integer_dtypes"], shape=xshape)
-        kw = array_api_kwargs(dtype=strategies.none() | self.onxps.scalar_dtypes())
+        kws = array_api_kwargs(dtype=strategies.none() | self.onxps.scalar_dtypes())
 
-        @given(x=xx, kw=kw)
+        @given(x=xx, kw=kws)
         def fctonx(x, kw):
+            asa = np.asarray(x.numpy())
+            try:
+                asp = onxp.asarray(x)
+            except Exception as e:
+                raise AssertionError(f"asarray fails with x={x!r}, asp={asa!r}.") from e
+            try:
+                self.assertEqualArray(asa, asp.numpy())
+            except AssertionError as e:
+                raise AssertionError(
+                    f"x={x!r} kw={kw!r} asa={asa!r}, asp={asp!r}"
+                ) from e
+            if kw:
+                try:
+                    asp2 = onxp.asarray(x, **kw)
+                except Exception as e:
+                    raise AssertionError(
+                        f"asarray fails with x={x!r}, kw={kw!r}, asp={asa!r}."
+                    ) from e
+                self.assertEqual(asp.shape, asp2.shape)
             args_onxp.append((x, kw))
 
         fctonx()

onnx_array_api/array_api/_onnx_common.py

@@ -1,5 +1,10 @@
 from typing import Any, Optional
+import warnings
 import numpy as np
+
+with warnings.catch_warnings():
+    warnings.simplefilter("ignore")
+    from numpy.array_api._array_object import Array
 from ..npx.npx_types import (
     DType,
     ElemType,
@@ -77,6 +82,10 @@ def asarray(
         v = TEagerTensor(np.array(a, dtype=np.str_))
     elif isinstance(a, list):
         v = TEagerTensor(np.array(a))
+    elif isinstance(a, np.ndarray):
+        v = TEagerTensor(a)
+    elif isinstance(a, Array):
+        v = TEagerTensor(np.asarray(a))
     else:
         raise RuntimeError(f"Unexpected type {type(a)} for the first input.")
     if dtype is not None:

onnx_array_api/npx/npx_numpy_tensors.py

@@ -1,3 +1,4 @@
+import warnings
 from typing import Any, Callable, List, Optional, Tuple
 import numpy as np
 from onnx import ModelProto, TensorProto
@@ -221,13 +222,18 @@ def __bool__(self):
         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}."
+            warnings.warn(
+                f"Conversion to bool only works for scalar, not for {self!r}, "
+                f"bool(...)={bool(self._tensor)}."
             )
+            try:
+                return bool(self._tensor)
+            except ValueError as e:
+                raise ValueError(f"Unable to convert {self} to bool.") from e
         return bool(self._tensor)
 
     def __int__(self):
-        "Implicit conversion to bool."
+        "Implicit conversion to int."
         if len(self.shape) != 0:
             raise ValueError(
                 f"Conversion to bool only works for scalar, not for {self!r}."
@@ -249,7 +255,7 @@ def __int__(self):
         return int(self._tensor)
 
     def __float__(self):
-        "Implicit conversion to bool."
+        "Implicit conversion to float."
         if len(self.shape) != 0:
             raise ValueError(
                 f"Conversion to bool only works for scalar, not for {self!r}."
@@ -261,11 +267,24 @@ def __float__(self):
             DType(TensorProto.BFLOAT16),
         }:
             raise TypeError(
-                f"Conversion to int only works for float scalar, "
+                f"Conversion to float only works for float scalar, "
                 f"not for dtype={self.dtype}."
             )
         return float(self._tensor)
 
+    def __iter__(self):
+        """
+        The :epkg:`Array API` does not define this function (2022/12).
+        This method raises an exception with a better error message.
+        """
+        warnings.warn(
+            f"Iterators are not implemented in the generic case. "
+            f"Every function using them cannot be converted into ONNX "
+            f"(tensors - {type(self)})."
+        )
+        for row in self._tensor:
+            yield self.__class__(row)
+
 
 class JitNumpyTensor(NumpyTensor, JitTensor):
     """

onnx_array_api/npx/npx_tensors.py

@@ -35,8 +35,9 @@ def __iter__(self):
         This method raises an exception with a better error message.
         """
         raise ArrayApiError(
-            "Iterators are not implemented in the generic case. "
-            "Every function using them cannot be converted into ONNX."
+            f"Iterators are not implemented in the generic case. "
+            f"Every function using them cannot be converted into ONNX "
+            f"(tensors - {type(self)})."
         )
 
     @staticmethod

onnx_array_api/npx/npx_types.py

@@ -59,12 +59,16 @@ def __eq__(self, dt: "DType") -> bool:
             return False
         if dt.__class__ is DType:
             return self.code_ == dt.code_
-        if isinstance(dt, (int, bool, str)):
+        if isinstance(dt, (int, bool, str, float)):
             return False
+        if dt is int:
+            return self.code_ == TensorProto.INT64
         if dt is str:
             return self.code_ == TensorProto.STRING
         if dt is bool:
             return self.code_ == TensorProto.BOOL
+        if dt is float:
+            return self.code_ == TensorProto.FLOAT64
         if isinstance(dt, list):
             return False
         if dt in ElemType.numpy_map:

onnx_array_api/npx/npx_var.py

@@ -607,8 +607,9 @@ def __iter__(self):
         This method raises an exception with a better error message.
         """
         raise ArrayApiError(
-            "Iterators are not implemented in the generic case. "
-            "Every function using them cannot be converted into ONNX."
+            f"Iterators are not implemented in the generic case. "
+            f"Every function using them cannot be converted into ONNX "
+            f"(Var - {type(self)})."
         )
 
     def _binary_op(self, ov: "Var", op_name: str, **kwargs) -> "Var":

requirements-dev.txt

@@ -3,7 +3,7 @@ black
 coverage
 flake8
 furo
-hypothesis<6.80.0
+hypothesis
 isort
 joblib
 lightgbm