add general -1 dim behavior for BroadcastTo op

mindspore/ops/operations/array_ops.py

@@ -4551,7 +4551,9 @@ class BroadcastTo(PrimitiveWithInfer):
     the target dimension is -1. In case of -1 in target shape, it will be replaced by the input shape's value
     in that dimension.
 
-    When input shape is broadcast to target shape, it starts with the trailing dimensions.
+    When input shape is broadcast to target shape, it starts with the trailing
+    dimensions. If there is a -1 in the target shape, the -1 cannot be in a leading,
+    non-existing dimension.
 
     Args:
         shape (tuple): The target shape to broadcast. Can be fully specified, or have -1 in one position
@@ -4566,9 +4568,8 @@ class BroadcastTo(PrimitiveWithInfer):
 
     Raises:
         TypeError: If `shape` is not a tuple.
-        ValueError: Given a shape tuple, if it has several -1; or if the -1 is in an invalid position
-            such as one that does not have a opposing dimension in an input tensor; or if the target and
-            input shapes are incompatible.
+        ValueError: if the target and input shapes are incompatible, or if a -1 in the
+        target shape is in an invalid location.
 
     Supported Platforms:
         ``Ascend`` ``GPU``
@@ -4582,13 +4583,13 @@ class BroadcastTo(PrimitiveWithInfer):
         [[1. 2. 3.]
          [1. 2. 3.]]
 
-        >>> shape = (2, -1)
-        >>> input_x = Tensor(np.array([1, 2, 3]).astype(np.float32))
+        >>> shape = (-1, 2)
+        >>> input_x = Tensor(np.array([[1], [2]]).astype(np.float32))
         >>> broadcast_to = ops.BroadcastTo(shape)
         >>> output = broadcast_to(input_x)
         >>> print(output)
-        [[1. 2. 3.]
-         [1. 2. 3.]]
+        [[1. 1.]
+         [2. 2.]]
     """
 
     @prim_attr_register
@@ -4600,35 +4601,30 @@ class BroadcastTo(PrimitiveWithInfer):
             validator.check_value_type('target shape index -> ' + str(ix), i, [int], self.name)
             validator.check("shape element", i, "shape element min limit", -1, Rel.GE, self.name)
         self.shape = shape
-        if -1 in self.shape:
-            undef_dims = self.shape.count(-1)
-            if undef_dims > 1:
-                raise ValueError(f'The shape can only has one -1 at most, but has {undef_dims}.')
-            self.dyn = True
-        else:
-            self.dyn = False
 
     def infer_shape(self, x_shape):
         validator.check("input_x shape length", len(x_shape), "target shape", len(self.shape), Rel.LE, self.name)
-        target_shape = list(self.shape)
-        outer_dim_offset = len(target_shape) - len(x_shape)
-        if self.dyn:
-            for i, v in enumerate(target_shape):
-                if v == -1:
-                    if i < outer_dim_offset:
-                        raise ValueError(f" -1 in init shape is in an incompatible location"
-                                         f" with given input tensor, -1 index in init shape: {i}"
-                                         f" but -1 can only be in index {len(x_shape)} onwards for this input.")
-                    target_shape[i] = x_shape[i - outer_dim_offset]
+
         reversed_x_shape = tuple(reversed(x_shape))
-        reversed_target = tuple(reversed(target_shape))
-        for i, v in enumerate(reversed_x_shape):
-            if v not in (reversed_target[i], 1):
-                raise ValueError(f"Not supported shapes for broadcast, "
-                                 f"x_shape: {tuple(x_shape)}, target shape {target_shape}.")
-        self.shape = tuple(target_shape)
+        reversed_filtered_target = []
+        for i, v in enumerate(tuple(reversed(self.shape))):
+            if v == -1:
+                if i >= len(reversed_x_shape):
+                    raise ValueError("-1 is not valid in a leading, non-existing dimension")
+
+                reversed_filtered_target.append(reversed_x_shape[i])
+            else:
+                reversed_filtered_target.append(v)
+
+        self.shape = tuple(reversed(reversed_filtered_target))
         self.add_prim_attr('shape', self.shape)
-        return target_shape
+
+        for i, v in enumerate(reversed_x_shape):
+            if v not in (reversed_filtered_target[i], 1):
+                raise ValueError(f"Not supported shapes for broadcast, "
+                                 f"x_shape: {tuple(x_shape)}, target shape {self.shape}.")
+
+        return self.shape
 
     def infer_dtype(self, x_dtype):
         validator.check_subclass("input_x", x_dtype, mstype.tensor, self.name)

tests/st/ops/gpu/test_broadcast_to_ops.py

@@ -1,4 +1,4 @@
-# Copyright 2020 Huawei Technologies Co., Ltd
+# Copyright 2020-2021 Huawei Technologies Co., Ltd
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -54,7 +54,7 @@ def test_broadcast_dyn_init():
     """
     context.set_context(mode=context.GRAPH_MODE, device_target='GPU')
 
-    ms_shape = (-1, 4, 5, 6)
+    ms_shape = (-1, -1, 5, 6)
     np_shape = (3, 4, 5, 6)
     x_np = np.random.rand(3, 1, 5, 1).astype(np.float32)
     output = P.BroadcastTo(ms_shape)(Tensor(x_np))
@@ -66,7 +66,7 @@ def test_broadcast_dyn_init():
     expect = np.broadcast_to(x1_np, np_shape)
     assert np.allclose(output.asnumpy(), expect)
 
-    ms_shape = (2, 3, -1, 5)
+    ms_shape = (2, 3, -1, -1)
     np_shape = (2, 3, 4, 5)
     x1_np = np.random.rand(4, 5).astype(np.float32)
     output = P.BroadcastTo(ms_shape)(Tensor(x1_np))
@@ -87,3 +87,9 @@ def test_broadcast_dyn_invalid_init():
     x_np = np.random.rand(4, 5).astype(np.float32)
     with pytest.raises(ValueError):
         P.BroadcastTo(ms_shape)(Tensor(x_np))
+
+    context.set_context(mode=context.GRAPH_MODE, device_target='GPU')
+    ms_shape = (-1, 1, -1, -1)
+    x_np = np.random.rand(4, 5).astype(np.float32)
+    with pytest.raises(ValueError):
+        P.BroadcastTo(ms_shape)(Tensor(x_np))