• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1# Copyright 2020 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.ops.operations import _grad_ops as G
23
24
25class NetEluGrad(nn.Cell):
26    def __init__(self):
27        super(NetEluGrad, self).__init__()
28        self.eluGrad = G.EluGrad()
29
30    def construct(self, x, dy):
31        return self.eluGrad(dy, x)
32
33
34@pytest.mark.level0
35@pytest.mark.platform_x86_gpu_training
36@pytest.mark.env_onecard
37def test_elu_grad_fp16():
38    x = Tensor(np.array([[0.5, 2, 5.5], [4.5, -2, 0]]).astype(np.float16))
39    dy = Tensor(np.array([[2, 1, 1.5], [-0.5, -1, -3]]).astype(np.float16))
40    expect = np.array([[2, 1, 1.5], [-0.5, 1, -3]]).astype(np.float16)
41    error = np.ones(shape=[2, 3]) * 1.0e-6
42
43    context.set_context(mode=context.GRAPH_MODE, device_target="GPU")
44    elu_grad = NetEluGrad()
45    output = elu_grad(x, dy)
46    diff = output.asnumpy() - expect
47    assert np.all(diff < error)
48
49@pytest.mark.level0
50@pytest.mark.platform_x86_gpu_training
51@pytest.mark.env_onecard
52def test_elu_grad_fp32():
53    x = Tensor(np.array([[0.5, 2, 5.5], [4.5, -2, 0]]).astype(np.float32))
54    dy = Tensor(np.array([[2, 1, 1.5], [-0.5, -1, -3]]).astype(np.float32))
55    expect = np.array([[2, 1, 1.5], [-0.5, 1, -3]]).astype(np.float32)
56    error = np.ones(shape=[2, 3]) * 1.0e-6
57
58    context.set_context(mode=context.GRAPH_MODE, device_target="GPU")
59    elu_grad = NetEluGrad()
60    output = elu_grad(x, dy)
61    diff = output.asnumpy() - expect
62    assert np.all(diff < error)
63