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""" 16test ssim 17""" 18import numpy as np 19import pytest 20 21import mindspore.common.dtype as mstype 22import mindspore.nn as nn 23from mindspore import Tensor 24from mindspore.common.api import _cell_graph_executor 25 26 27class SSIMNet(nn.Cell): 28 def __init__(self, max_val=1.0, filter_size=11, filter_sigma=1.5, k1=0.01, k2=0.03): 29 super(SSIMNet, self).__init__() 30 self.net = nn.SSIM(max_val, filter_size, filter_sigma, k1, k2) 31 32 def construct(self, img1, img2): 33 return self.net(img1, img2) 34 35 36def test_compile(): 37 net = SSIMNet() 38 img1 = Tensor(np.random.random((8, 3, 16, 16)), mstype.float32) 39 img2 = Tensor(np.random.random((8, 3, 16, 16)), mstype.float32) 40 _cell_graph_executor.compile(net, img1, img2) 41 42 43def test_ssim_max_val_negative(): 44 max_val = -1 45 with pytest.raises(ValueError): 46 _ = SSIMNet(max_val) 47 48 49def test_ssim_max_val_bool(): 50 max_val = True 51 with pytest.raises(TypeError): 52 _ = SSIMNet(max_val) 53 54 55def test_ssim_max_val_zero(): 56 max_val = 0 57 with pytest.raises(ValueError): 58 _ = SSIMNet(max_val) 59 60 61def test_ssim_filter_size_float(): 62 with pytest.raises(TypeError): 63 _ = SSIMNet(filter_size=1.1) 64 65 66def test_ssim_filter_size_zero(): 67 with pytest.raises(ValueError): 68 _ = SSIMNet(filter_size=0) 69 70 71def test_ssim_filter_sigma_zero(): 72 with pytest.raises(ValueError): 73 _ = SSIMNet(filter_sigma=0.0) 74 75 76def test_ssim_filter_sigma_negative(): 77 with pytest.raises(ValueError): 78 _ = SSIMNet(filter_sigma=-0.1) 79 80 81def test_ssim_different_shape(): 82 shape_1 = (8, 3, 16, 16) 83 shape_2 = (8, 3, 8, 8) 84 img1 = Tensor(np.random.random(shape_1)) 85 img2 = Tensor(np.random.random(shape_2)) 86 net = SSIMNet() 87 with pytest.raises(TypeError): 88 _cell_graph_executor.compile(net, img1, img2) 89 90 91def test_ssim_different_dtype(): 92 dtype_1 = mstype.float32 93 dtype_2 = mstype.float16 94 img1 = Tensor(np.random.random((8, 3, 16, 16)), dtype=dtype_1) 95 img2 = Tensor(np.random.random((8, 3, 16, 16)), dtype=dtype_2) 96 net = SSIMNet() 97 with pytest.raises(TypeError): 98 _cell_graph_executor.compile(net, img1, img2) 99 100 101def test_ssim_invalid_5d_input(): 102 shape_1 = (8, 3, 16, 16) 103 shape_2 = (8, 3, 8, 8) 104 invalid_shape = (8, 3, 16, 16, 1) 105 img1 = Tensor(np.random.random(shape_1)) 106 invalid_img1 = Tensor(np.random.random(invalid_shape)) 107 img2 = Tensor(np.random.random(shape_2)) 108 invalid_img2 = Tensor(np.random.random(invalid_shape)) 109 110 net = SSIMNet() 111 with pytest.raises(TypeError): 112 _cell_graph_executor.compile(net, invalid_img1, img2) 113 with pytest.raises(TypeError): 114 _cell_graph_executor.compile(net, img1, invalid_img2) 115 with pytest.raises(TypeError): 116 _cell_graph_executor.compile(net, invalid_img1, invalid_img2) 117