!1363 Restrict tensor get item or set item not support mixed tensor.

Merge pull request !1363 from zhangbuxue/Not_support_mixed_tensor_as_index_for_tensor_getitem_and_setitem

mindspore/ops/composite/multitype_ops/_utils.py

@@ -254,7 +254,7 @@ def tuple_element_is_int(indexs):
 
 
 @constexpr
-def tuple_elements_type(types):
+def tuple_index_elements_type(types, op_name):
     """Judges the type of all elements of the tuple."""
     tensors_number = 0
     for ele in types:
@@ -264,7 +264,7 @@ def tuple_elements_type(types):
         return ALL_TENSOR
     if tensors_number == 0:
         return NO_TENSOR
-    return CONTAIN_TENSOR
+    raise IndexError(f"For '{op_name}', the index does not support mixed tensor.")
 
 
 @constexpr

mindspore/ops/composite/multitype_ops/getitem_impl.py

@@ -247,7 +247,7 @@ def _tensor_getitem_by_tuple(data, tuple_index):
         Tensor, element type is same as the element type of data.
     """
     index_types = multi_utils.hyper_map(F.typeof, tuple_index)
-    index_elements_type = multi_utils.tuple_elements_type(index_types)
+    index_elements_type = multi_utils.tuple_index_elements_type(index_types, multi_utils.TENSOR_GETITEM)
     result = None
     if index_elements_type == multi_utils.NO_TENSOR:
         result = _tensor_slice(data, tuple_index)

mindspore/ops/composite/multitype_ops/setitem_impl.py

@@ -191,7 +191,7 @@ def _tensor_setitem_by_tuple_with_number(data, tuple_index, value):
         Tensor, element type and shape is same as data.
     """
     index_types = multi_utils.hyper_map(F.typeof, tuple_index)
-    index_elements_type = multi_utils.tuple_elements_type(index_types)
+    index_elements_type = multi_utils.tuple_index_elements_type(index_types, multi_utils.TENSOR_SETITEM)
     result = None
     if index_elements_type == multi_utils.NO_TENSOR:
         result = _tensor_assgin_number(data, tuple_index, value)
@@ -222,7 +222,7 @@ def _tensor_setitem_by_tuple_with_tensor(data, tuple_index, value):
         Tensor, element type and shape is same as data.
     """
     index_types = multi_utils.hyper_map(F.typeof, tuple_index)
-    index_elements_type = multi_utils.tuple_elements_type(index_types)
+    index_elements_type = multi_utils.tuple_index_elements_type(index_types, multi_utils.TENSOR_SETITEM)
     result = None
     if index_elements_type == multi_utils.NO_TENSOR:
         result = _tensor_assgin_tensor(data, tuple_index, value)
@@ -254,7 +254,7 @@ def _tensor_setitem_by_tuple_with_tuple(data, tuple_index, value):
         Tensor, element type and shape is same as data.
     """
     index_types = multi_utils.hyper_map(F.typeof, tuple_index)
-    index_elements_type = multi_utils.tuple_elements_type(index_types)
+    index_elements_type = multi_utils.tuple_index_elements_type(index_types, multi_utils.TENSOR_SETITEM)
     result = None
     if index_elements_type == multi_utils.ALL_TENSOR:
         indices = multi_utils.generate_indeices_from_tuple_of_tensor(data, tuple_index, multi_utils.TENSOR_SETITEM)

tests/ut/python/ops/test_tensor_slice.py

@@ -146,9 +146,9 @@ class TensorAssignWithSlice(Cell):
         return z
 
 
-class TensorIndexByOneTensor(Cell):
+class TensorGetItemByOneTensor(Cell):
     def __init__(self):
-        super(TensorIndexByOneTensor, self).__init__()
+        super(TensorGetItemByOneTensor, self).__init__()
         self.const = Tensor(np.ones((5, 4, 7, 8)), mstype.int32)
 
     def construct(self, x, index):
@@ -156,9 +156,9 @@ class TensorIndexByOneTensor(Cell):
         return ret
 
 
-class TensorIndexByTwoTensors(Cell):
+class TensorGetItemByTwoTensors(Cell):
     def __init__(self):
-        super(TensorIndexByTwoTensors, self).__init__()
+        super(TensorGetItemByTwoTensors, self).__init__()
         self.const = Tensor(np.ones((3, 4, 5, 8)), mstype.int32)
 
     def construct(self, x, index_0, index_1):
@@ -166,9 +166,9 @@ class TensorIndexByTwoTensors(Cell):
         return ret
 
 
-class TensorIndexByThreeTensors(Cell):
+class TensorGetItemByThreeTensors(Cell):
     def __init__(self):
-        super(TensorIndexByThreeTensors, self).__init__()
+        super(TensorGetItemByThreeTensors, self).__init__()
         self.const = Tensor(np.ones((5, 3, 4, 5)), mstype.int32)
 
     def construct(self, x, index_0, index_1, index_2):
@@ -176,6 +176,15 @@ class TensorIndexByThreeTensors(Cell):
         return ret
 
 
+class TensorGetItemByMixedTensors(Cell):
+    def __init__(self):
+        super(TensorGetItemByMixedTensors, self).__init__()
+
+    def construct(self, x, index_0, index_1):
+        ret = x[index_0, index_1, 0:6]
+        return ret
+
+
 class TensorSetItemByOneTensorWithNumber(Cell):
     def __init__(self, value):
         super(TensorSetItemByOneTensorWithNumber, self).__init__()
@@ -300,6 +309,19 @@ class TensorSetItemByTensorsWithTupleOfTensorNumberError(Cell):
         return ret
 
 
+class TensorSetItemByMixedTensors(Cell):
+    def __init__(self):
+        super(TensorSetItemByMixedTensors, self).__init__()
+        self.const = Tensor(np.ones((6, 7, 8)), mstype.float32)
+        self.param = Parameter(Tensor(np.arange(6*7*8).reshape((6, 7, 8)), mstype.float32), name="x")
+        self.value = 99.0
+
+    def construct(self, index_0, index_1):
+        self.param[index_0, index_1, 0:6] = self.value
+        ret = self.param + self.const
+        return ret
+
+
 def test_tensor_assign():
     context.set_context(mode=context.GRAPH_MODE, save_graphs=True)
     net = TensorAssignWithSlice()
@@ -596,19 +618,19 @@ test_cases = [
         'block': NetWorkSliceEllipsis(),
         'desc_inputs': [Tensor(np.ones([6, 7, 8, 9], np.int32))],
     }),
-    ('TensorIndexByOneTensor', {
-        'block': TensorIndexByOneTensor(),
+    ('TensorGetItemByOneTensor', {
+        'block': TensorGetItemByOneTensor(),
         'desc_inputs': [Tensor(np.arange(6*7*8).reshape((6, 7, 8)), mstype.int32),
                         Tensor(np.random.randint(6, size=(5, 4)), mstype.int32)],
     }),
-    ('TensorIndexByTwoTensors', {
-        'block': TensorIndexByTwoTensors(),
+    ('TensorGetItemByTwoTensors', {
+        'block': TensorGetItemByTwoTensors(),
         'desc_inputs': [Tensor(np.arange(6*7*8).reshape((6, 7, 8)), mstype.int32),
                         Tensor(np.random.randint(6, size=(3, 4, 5)), mstype.int32),
                         Tensor(np.random.randint(7, size=(4, 5)), mstype.int32)],
     }),
-    ('TensorIndexByThreeTensors', {
-        'block': TensorIndexByThreeTensors(),
+    ('TensorGetItemByThreeTensors', {
+        'block': TensorGetItemByThreeTensors(),
         'desc_inputs': [Tensor(np.arange(6*7*8).reshape((6, 7, 8)), mstype.int32),
                         Tensor(np.random.randint(6, size=(3, 4, 5)), mstype.int32),
                         Tensor(np.random.randint(7, size=(4, 5)), mstype.int32),
@@ -665,37 +687,43 @@ test_cases = [
 ]
 
 raise_error_set = [
-    ('TensorIndexByOneTensorDtypeError', {
-        'block': (TensorIndexByOneTensor(), {'exception': TypeError}),
+    ('TensorGetItemByOneTensorDtypeError', {
+        'block': (TensorGetItemByOneTensor(), {'exception': TypeError}),
         'desc_inputs': [Tensor(np.arange(6*7*8).reshape((6, 7, 8)), mstype.int32),
                         Tensor(np.random.randint(6, size=(5, 4)), mstype.int8)],
     }),
-    ('TensorIndexByTwoTensorsShapeError', {
-        'block': (TensorIndexByTwoTensors(), {'exception': ValueError}),
+    ('TensorGetItemByTwoTensorsShapeError', {
+        'block': (TensorGetItemByTwoTensors(), {'exception': ValueError}),
         'desc_inputs': [Tensor(np.arange(6*7*8).reshape((6, 7, 8)), mstype.int32),
                         Tensor(np.random.randint(6, size=(3, 4, 5)), mstype.int32),
                         Tensor(np.random.randint(7, size=(2, 3, 5)), mstype.int32)],
     }),
-    ('TensorIndexByTwoTensorsDtypeError', {
-        'block': (TensorIndexByTwoTensors(), {'exception': TypeError}),
+    ('TensorGetItemByTwoTensorsDtypeError', {
+        'block': (TensorGetItemByTwoTensors(), {'exception': TypeError}),
         'desc_inputs': [Tensor(np.arange(6*7*8).reshape((6, 7, 8)), mstype.int32),
                         Tensor(np.random.randint(6, size=(3, 4, 5)), mstype.int32),
                         Tensor(np.random.randint(7, size=(4, 5)), mstype.float32)],
     }),
-    ('TensorIndexByThreeTensorsShapeError', {
-        'block': (TensorIndexByThreeTensors(), {'exception': ValueError}),
+    ('TensorGetItemByThreeTensorsShapeError', {
+        'block': (TensorGetItemByThreeTensors(), {'exception': ValueError}),
         'desc_inputs': [Tensor(np.arange(6*7*8).reshape((6, 7, 8)), mstype.int32),
                         Tensor(np.random.randint(6, size=(3, 4, 5)), mstype.int32),
                         Tensor(np.random.randint(7, size=(3, 4, 5)), mstype.int32),
                         Tensor(np.random.randint(8, size=(5, 2, 4, 5)), mstype.int32)],
     }),
-    ('TensorIndexByThreeTensorsDtypeError', {
-        'block': (TensorIndexByThreeTensors(), {'exception': TypeError}),
+    ('TensorGetItemByThreeTensorsDtypeError', {
+        'block': (TensorGetItemByThreeTensors(), {'exception': TypeError}),
         'desc_inputs': [Tensor(np.arange(6*7*8).reshape((6, 7, 8)), mstype.int32),
                         Tensor(np.random.randint(6, size=(3, 4, 5)), mstype.int32),
                         Tensor(np.random.randint(7, size=(3, 4, 5)), mstype.int64),
                         Tensor(np.random.randint(8, size=(5, 3, 4, 5)), mstype.int32)],
     }),
+    ('TensorGetItemByMixedTensors', {
+        'block': (TensorGetItemByMixedTensors(), {'exception': IndexError}),
+        'desc_inputs': [Tensor(np.arange(6*7*8).reshape((6, 7, 8)), mstype.int32),
+                        Tensor(np.random.randint(6, size=(3, 4, 5)), mstype.int32),
+                        Tensor(np.random.randint(7, size=(3, 4, 5)), mstype.int64)],
+    }),
     ('TensorSetItemByOneTensorWithNumberTypeError', {
         'block': (TensorSetItemByOneTensorWithNumber(value=0), {'exception': TypeError}),
         'desc_inputs': [Tensor(np.random.randint(4, size=(5, 4)), mstype.int32)],
@@ -781,6 +809,11 @@ raise_error_set = [
                         Tensor(np.zeros((4, 5)), mstype.float32),
                         Tensor(np.ones((4, 5)), mstype.int32),
                         Tensor(np.ones((4, 5)) * 2, mstype.int32)],
+    }),
+    ('TensorSetItemByMixedTensors', {
+        'block': (TensorSetItemByMixedTensors(),  {'exception': IndexError}),
+        'desc_inputs': [Tensor(np.random.randint(6, size=(3, 4, 5)), mstype.int32),
+                        Tensor(np.random.randint(7, size=(4, 5)), mstype.int32)],
     })
 ]