[NPU] Support npu kernel for atan and atan_grad op, test=develop (#34658)

* fix npu compile error, test=develop

* [NPU] Support npu kernel for atan and atan_grad op, test=develop

* [NPU] Support npu kernel for atan and atan_grad op, test=develop
Co-authored-by: Nqili93 <qili93@qq.com>

paddle/fluid/operators/activation_op_npu.cc

@@ -527,6 +527,39 @@ class CosGradNPUKernel : public framework::OpKernel<T> {
   }
 };
 
+template <typename DeviceContext, typename T>
+class AtanNPUKernel : public framework::OpKernel<T> {
+ public:
+  void Compute(const framework::ExecutionContext& ctx) const override {
+    auto* x = ctx.Input<Tensor>("X");
+    auto* out = ctx.Output<Tensor>("Out");
+    auto place = ctx.GetPlace();
+    out->mutable_data<T>(place);
+    const auto& runner = NpuOpRunner("Atan", {*x}, {*out}, {});
+    auto stream =
+        ctx.template device_context<paddle::platform::NPUDeviceContext>()
+            .stream();
+    runner.Run(stream);
+  }
+};
+
+template <typename DeviceContext, typename T>
+class AtanGradNPUKernel : public framework::OpKernel<T> {
+ public:
+  void Compute(const framework::ExecutionContext& ctx) const override {
+    auto* dout = ctx.Input<Tensor>(framework::GradVarName("Out"));
+    auto* x = ctx.Input<Tensor>("X");
+    auto* dx = ctx.Output<Tensor>(framework::GradVarName("X"));
+    auto place = ctx.GetPlace();
+    dx->mutable_data<T>(place);
+    auto stream =
+        ctx.template device_context<paddle::platform::NPUDeviceContext>()
+            .stream();
+    const auto& runner_dx = NpuOpRunner("AtanGrad", {*x, *dout}, {*dx}, {});
+    runner_dx.Run(stream);
+  }
+};
+
 }  // namespace operators
 }  // namespace paddle
 
@@ -648,3 +681,14 @@ REGISTER_OP_NPU_KERNEL(
     cos_grad, ops::CosGradNPUKernel<paddle::platform::NPUDeviceContext, float>,
     ops::CosGradNPUKernel<paddle::platform::NPUDeviceContext,
                           paddle::platform::float16>);
+
+REGISTER_OP_NPU_KERNEL(
+    atan, ops::AtanNPUKernel<paddle::platform::NPUDeviceContext, float>,
+    ops::AtanNPUKernel<paddle::platform::NPUDeviceContext,
+                       paddle::platform::float16>);
+
+REGISTER_OP_NPU_KERNEL(
+    atan_grad,
+    ops::AtanGradNPUKernel<paddle::platform::NPUDeviceContext, float>,
+    ops::AtanGradNPUKernel<paddle::platform::NPUDeviceContext,
+                           paddle::platform::float16>);

python/paddle/fluid/tests/unittests/npu/test_atan_op_npu.py

+#  Copyright (c) 2021 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.
+
+from __future__ import print_function
+
+import numpy as np
+import unittest
+import sys
+sys.path.append("..")
+from op_test import OpTest
+import paddle
+import paddle.fluid as fluid
+
+paddle.enable_static()
+SEED = 1024
+
+
+class TestAtan(OpTest):
+    def setUp(self):
+        self.set_npu()
+        self.op_type = "atan"
+        self.place = paddle.NPUPlace(0)
+
+        self.dtype = np.float32
+        np.random.seed(SEED)
+        self.shape = [11, 17]
+        x = np.random.uniform(0.1, 1, self.shape).astype(self.dtype)
+        out = np.arctan(x)
+
+        self.inputs = {'X': OpTest.np_dtype_to_fluid_dtype(x)}
+        self.outputs = {'Out': out}
+
+    def set_attrs(self):
+        pass
+
+    def set_npu(self):
+        self.__class__.use_npu = True
+
+    def test_check_grad(self):
+        self.check_grad_with_place(self.place, ['X'], 'Out')
+
+    def test_out_name(self):
+        with fluid.program_guard(fluid.Program()):
+            np_x = np.array([0.1])
+            data = fluid.layers.data(name="X", shape=[1])
+            out = paddle.atan(data, name='Y')
+            place = paddle.NPUPlace(0)
+            exe = fluid.Executor(place)
+            result, = exe.run(feed={"X": np_x}, fetch_list=[out])
+            expected = np.arctan(np_x)
+            self.assertEqual(result, expected)
+
+    def test_dygraph(self):
+        with fluid.dygraph.guard(paddle.NPUPlace(0)):
+            np_x = np.array([0.1])
+            x = fluid.dygraph.to_variable(np_x)
+            z = paddle.atan(x).numpy()
+            z_expected = np.arctan(np_x)
+            self.assertEqual(z, z_expected)
+
+    def test_check_output(self):
+        self.check_output_with_place(self.place)
+
+
+class TestAtanShape(TestAtan):
+    def set_attrs(self):
+        self.shape = [12, 23, 10]
+
+
+class TestAtanFloat16(TestAtan):
+    def set_attrs(self):
+        self.dtype = np.float16
+
+
+if __name__ == '__main__':
+    unittest.main()