【Hackathon 5th No.45】为torch.cuda.comm.scatter and torch.cuda.comm.gat…

…her 添加test (#319)

add test for cuda_comm_scatter and cuda_comm_gather

tests/test_cuda_comm_gather.py

@@ -0,0 +1,71 @@
+# Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+import textwrap
+
+from apibase import APIBase
+
+obj = APIBase("torch.cuda.comm.gather")
+
+
+def test_case_1():
+    pytorch_code = textwrap.dedent(
+        """
+        import torch
+        t1 = torch.randn((10, 3, 32, 32), device='cpu')
+        t2 = torch.randn((10, 3, 32, 32), device='cpu')
+        results = torch.cuda.comm.gather([t1, t2])
+        """
+    )
+    obj.run(
+        pytorch_code,
+        ["result"],
+        unsupport=True,
+        reason="paddle has no corresponding api tentatively",
+    )
+
+
+def test_case_2():
+    pytorch_code = textwrap.dedent(
+        """
+        import torch
+        t1 = torch.randn((10, 3, 32, 32), device='cpu')
+        t2 = torch.randn((10, 3, 32, 32), device='cpu')
+        results = torch.cuda.comm.gather([t1, t2], dims=1)
+        """
+    )
+    obj.run(
+        pytorch_code,
+        ["result"],
+        unsupport=True,
+        reason="paddle has no corresponding api tentatively",
+    )
+
+
+def test_case_3():
+    pytorch_code = textwrap.dedent(
+        """
+        import torch
+        t1 = torch.randn((10, 3, 32, 32), device='cpu')
+        t2 = torch.randn((10, 3, 32, 32), device='cpu')
+        results = torch.cuda.comm.gather([t1, t2], destination='cpu')
+        """
+    )
+    obj.run(
+        pytorch_code,
+        ["result"],
+        unsupport=True,
+        reason="paddle has no corresponding api tentatively",
+    )

tests/test_cuda_comm_scatter.py

@@ -0,0 +1,124 @@
+# Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+import textwrap
+
+from apibase import APIBase
+
+obj = APIBase("torch.cuda.comm.scatter")
+
+
+def test_case_1():
+    pytorch_code = textwrap.dedent(
+        """
+        import torch
+        nhwc = torch.randn((10, 3, 32, 32), device='cpu')
+        results = torch.cuda.comm.scatter(tensor=nhwc)
+        """
+    )
+    obj.run(
+        pytorch_code,
+        ["result"],
+        unsupport=True,
+        reason="paddle has no corresponding api tentatively",
+    )
+
+
+def test_case_2():
+    pytorch_code = textwrap.dedent(
+        """
+        import torch
+        nhwc = torch.randn((10, 3, 32, 32), device='gpu')
+        results = torch.cuda.comm.scatter(tensor=nhwc)
+        """
+    )
+    obj.run(
+        pytorch_code,
+        ["result"],
+        unsupport=True,
+        reason="paddle has no corresponding api tentatively",
+    )
+
+
+def test_case_3():
+    pytorch_code = textwrap.dedent(
+        """
+        import torch
+        nhwc = torch.randn((10, 3, 32, 32), device='gpu')
+        devices = [torch.device('cuda:0'), torch.device('cuda:1')]
+        result = torch.cuda.comm.scatter(nhwc, devices=devices)
+    """
+    )
+
+    obj.run(
+        pytorch_code,
+        ["result"],
+        unsupport=True,
+        reason="paddle has no corresponding api tentatively",
+    )
+
+
+def test_case_4():
+    pytorch_code = textwrap.dedent(
+        """
+        import torch
+        nhwc = torch.randn((10, 3, 32, 32), device='gpu')
+        chunk_sizes = [5, 5]
+        result = torch.cuda.comm.scatter(nhwc, chunk_sizes=chunk_sizes)
+    """
+    )
+
+    obj.run(
+        pytorch_code,
+        ["result"],
+        unsupport=True,
+        reason="paddle has no corresponding api tentatively",
+    )
+
+
+def test_case_5():
+    pytorch_code = textwrap.dedent(
+        """
+        import torch
+        nhwc = torch.randn((10, 3, 32, 32), device='gpu')
+        result = torch.cuda.comm.scatter(nhwc, dim=1)
+    """
+    )
+
+    obj.run(
+        pytorch_code,
+        ["result"],
+        unsupport=True,
+        reason="paddle has no corresponding api tentatively",
+    )
+
+
+def test_case_6():
+    pytorch_code = textwrap.dedent(
+        """
+        import torch
+        nhwc = torch.randn((10, 3, 32, 32), device='gpu')
+        t1 = torch.empty(5, 10, device='cuda')
+        t2 = torch.empty(5, 10, device='cuda')
+        result = torch.cuda.comm.scatter(nhwc, out=[t1, t2])
+    """
+    )
+
+    obj.run(
+        pytorch_code,
+        ["result"],
+        unsupport=True,
+        reason="paddle has no corresponding api tentatively",
+    )