• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1# Copyright 2020-2021 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.ops import operations as P
23
24class NetDiv(nn.Cell):
25    def __init__(self):
26        super(NetDiv, self).__init__()
27        self.div = P.Div()
28
29    def construct(self, x, y):
30        return self.div(x, y)
31
32def div(nptype):
33    x0_np = np.random.randint(1, 5, (2, 3, 4, 4)).astype(nptype)
34    y0_np = np.random.randint(1, 5, (2, 3, 4, 4)).astype(nptype)
35    x1_np = np.random.randint(1, 5, (2, 3, 4, 4)).astype(nptype)
36    y1_np = np.random.randint(1, 5, (2, 1, 4, 4)).astype(nptype)
37    x2_np = np.random.randint(1, 5, (2, 1, 1, 4)).astype(nptype)
38    y2_np = np.random.randint(1, 5, (2, 3, 4, 4)).astype(nptype)
39    x3_np = np.random.randint(1, 5, 1).astype(nptype)
40    y3_np = np.random.randint(1, 5, 1).astype(nptype)
41    x4_np = np.array(78).astype(nptype)
42    y4_np = np.array(37.5).astype(nptype)
43
44    x0 = Tensor(x0_np)
45    y0 = Tensor(y0_np)
46    x1 = Tensor(x1_np)
47    y1 = Tensor(y1_np)
48    x2 = Tensor(x2_np)
49    y2 = Tensor(y2_np)
50    x3 = Tensor(x3_np)
51    y3 = Tensor(y3_np)
52    x4 = Tensor(x4_np)
53    y4 = Tensor(y4_np)
54
55    context.set_context(mode=context.GRAPH_MODE, device_target='GPU')
56    div_net = NetDiv()
57    output0 = div_net(x0, y0)
58    expect0 = np.divide(x0_np, y0_np)
59    diff0 = output0.asnumpy() - expect0
60    error0 = np.ones(shape=expect0.shape) * 1.0e-5
61    assert np.all(diff0 < error0)
62    assert output0.shape == expect0.shape
63
64    output1 = div_net(x1, y1)
65    expect1 = np.divide(x1_np, y1_np)
66    diff1 = output1.asnumpy() - expect1
67    error1 = np.ones(shape=expect1.shape) * 1.0e-5
68    assert np.all(diff1 < error1)
69    assert output1.shape == expect1.shape
70
71    output2 = div_net(x2, y2)
72    expect2 = np.divide(x2_np, y2_np)
73    diff2 = output2.asnumpy() - expect2
74    error2 = np.ones(shape=expect2.shape) * 1.0e-5
75    assert np.all(diff2 < error2)
76    assert output2.shape == expect2.shape
77
78    context.set_context(mode=context.PYNATIVE_MODE, device_target='GPU')
79    output3 = div_net(x3, y3)
80    expect3 = np.divide(x3_np, y3_np)
81    diff3 = output3.asnumpy() - expect3
82    error3 = np.ones(shape=expect3.shape) * 1.0e-5
83    assert np.all(diff3 < error3)
84    assert output3.shape == expect3.shape
85
86    output4 = div_net(x4, y4)
87    expect4 = np.divide(x4_np, y4_np)
88    diff4 = output4.asnumpy() - expect4
89    error4 = np.ones(shape=expect4.shape) * 1.0e-5
90    assert np.all(diff4 < error4)
91    assert output4.shape == expect4.shape
92
93@pytest.mark.level0
94@pytest.mark.platform_x86_gpu_training
95@pytest.mark.env_onecard
96def test_div_float64():
97    div(np.float64)
98
99@pytest.mark.level0
100@pytest.mark.platform_x86_gpu_training
101@pytest.mark.env_onecard
102def test_div_float32():
103    div(np.float32)
104
105@pytest.mark.level1
106@pytest.mark.platform_x86_gpu_training
107@pytest.mark.env_onecard
108def test_div_float16():
109    div(np.float16)
110
111@pytest.mark.level1
112@pytest.mark.platform_x86_gpu_training
113@pytest.mark.env_onecard
114def test_div_int64():
115    div(np.int64)
116
117@pytest.mark.level1
118@pytest.mark.platform_x86_gpu_training
119@pytest.mark.env_onecard
120def test_div_int32():
121    div(np.int32)
122