• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1# Copyright 2021 Huawei Technologies Co., Ltd
2#
3# Licensed under the Apache License, Version 2.0 (the "License");
4# you may not use this file except in compliance with the License.
5# You may obtain a copy of the License at
6#
7# http://www.apache.org/licenses/LICENSE-2.0
8#
9# Unless required by applicable law or agreed to in writing, software
10# distributed under the License is distributed on an "AS IS" BASIS,
11# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12# See the License for the specific language governing permissions and
13# limitations under the License.
14# ============================================================================
15
16import numpy as np
17import pytest
18
19import mindspore.context as context
20import mindspore.nn as nn
21from mindspore import Tensor
22from mindspore.common.initializer import initializer
23from mindspore.common.parameter import Parameter
24from mindspore.ops.operations import _grad_ops as G
25
26context.set_context(mode=context.GRAPH_MODE, device_target='CPU')
27
28
29class NetReluGrad(nn.Cell):
30    def __init__(self):
31        super(NetReluGrad, self).__init__()
32        self.relu6_grad = G.ReLU6Grad()
33        self.x = Parameter(initializer(Tensor(np.array([[[[1, 0, 6],
34                                                          [-2, 3, 6],
35                                                          [-3, 1, 8]]]]).astype(np.float32)), [1, 1, 3, 3]), name='x')
36        self.dy = Parameter(initializer(Tensor(np.array([[[[1, 2, 3],
37                                                           [4, 5, 6],
38                                                           [7, 8, 9]]]]).astype(np.float32)), [1, 1, 3, 3]), name='dy')
39
40    def construct(self):
41        return self.relu6_grad(self.dy, self.x)
42
43
44@pytest.mark.level0
45@pytest.mark.platform_x86_cpu
46@pytest.mark.env_onecard
47def test_relu_grad():
48    relu_grad = NetReluGrad()
49    output = relu_grad()
50    expect = np.array([[[[1, 0, 3], [0, 5, 6], [0, 8, 0]]]]).astype(np.float32)
51    error = np.ones(shape=[3, 3]) * 1.0e-6
52    diff = np.abs(output.asnumpy() - expect)
53    assert np.all(diff < error)
54