!383 add cell psnr

Merge pull request !383 from zhaozhenlong/cell/psnr

mindspore/nn/layer/__init__.py

@@ -25,7 +25,7 @@ from .lstm import LSTM
 from .basic import Dropout, Flatten, Dense, ClipByNorm, Norm, OneHot, Pad, Unfold
 from .embedding import Embedding
 from .pooling import AvgPool2d, MaxPool2d
-from .image import ImageGradients, SSIM
+from .image import ImageGradients, SSIM, PSNR
 
 __all__ = ['Softmax', 'LogSoftmax', 'ReLU', 'ReLU6', 'Tanh', 'GELU', 'Sigmoid',
            'PReLU', 'get_activation', 'LeakyReLU', 'HSigmoid', 'HSwish', 'ELU',
@@ -36,5 +36,5 @@ __all__ = ['Softmax', 'LogSoftmax', 'ReLU', 'ReLU6', 'Tanh', 'GELU', 'Sigmoid',
            'Dropout', 'Flatten', 'Dense', 'ClipByNorm', 'Norm', 'OneHot',
            'Embedding',
            'AvgPool2d', 'MaxPool2d', 'Pad', 'Unfold',
-           'ImageGradients', 'SSIM',
+           'ImageGradients', 'SSIM', 'PSNR',
            ]

mindspore/nn/layer/image.py

@@ -69,6 +69,18 @@ class ImageGradients(Cell):
         return dy, dx
 
 
+def _convert_img_dtype_to_float32(img, max_val):
+    """convert img dtype to float32"""
+    # Ususally max_val is 1.0 or 255, we will do the scaling if max_val > 1.
+    # We will scale img pixel value if max_val > 1. and just cast otherwise.
+    ret = F.cast(img, mstype.float32)
+    max_val = F.scalar_cast(max_val, mstype.float32)
+    if max_val > 1.:
+        scale = 1. / max_val
+        ret = ret * scale
+    return ret
+
+
 @constexpr
 def _gauss_kernel_helper(filter_size):
     """gauss kernel helper"""
@@ -134,9 +146,9 @@ class SSIM(Cell):
         self.mean = P.DepthwiseConv2dNative(channel_multiplier=1, kernel_size=filter_size)
 
     def construct(self, img1, img2):
-        max_val = self._convert_img_dtype_to_float32(self.max_val, self.max_val)
-        img1 = self._convert_img_dtype_to_float32(img1, self.max_val)
-        img2 = self._convert_img_dtype_to_float32(img2, self.max_val)
+        max_val = _convert_img_dtype_to_float32(self.max_val, self.max_val)
+        img1 = _convert_img_dtype_to_float32(img1, self.max_val)
+        img2 = _convert_img_dtype_to_float32(img2, self.max_val)
 
         kernel = self._fspecial_gauss(self.filter_size, self.filter_sigma)
         kernel = P.Tile()(kernel, (1, P.Shape()(img1)[1], 1, 1))
@@ -145,21 +157,10 @@ class SSIM(Cell):
 
         return mean_ssim
 
-    def _convert_img_dtype_to_float32(self, img, max_val):
-        """convert img dtype to float32"""
-        # Ususally max_val is 1.0 or 255, we will do the scaling if max_val > 1.
-        # We will scale img pixel value if max_val > 1. and just cast otherwise.
-        ret = P.Cast()(img, mstype.float32)
-        max_val = F.scalar_cast(max_val, mstype.float32)
-        if max_val > 1.:
-            scale = 1./max_val
-            ret = ret*scale
-        return ret
-
     def _calculate_mean_ssim(self, x, y, kernel, max_val, k1, k2):
         """calculate mean ssim"""
-        c1 = (k1*max_val)*(k1*max_val)
-        c2 = (k2*max_val)*(k2*max_val)
+        c1 = (k1 * max_val) * (k1 * max_val)
+        c2 = (k2 * max_val) * (k2 * max_val)
 
         # SSIM luminance formula
         # (2 * mean_{x} * mean_{y} + c1) / (mean_{x}**2 + mean_{y}**2 + c1)
@@ -195,3 +196,52 @@ class SSIM(Cell):
         g = P.Softmax()(g)
         ret = F.reshape(g, (1, 1, filter_size, filter_size))
         return ret
+
+
+class PSNR(Cell):
+    r"""
+    Returns Peak Signal-to-Noise Ratio of two image batches.
+
+    It produces a PSNR value for each image in batch.
+    Assume inputs are :math:`I` and :math:`K`, both with shape :math:`h*w`.
+    :math:`MAX` represents the dynamic range of pixel values.
+
+    .. math::
+
+        MSE&=\frac{1}{hw}\sum\limits_{i=0}^{h-1}\sum\limits_{j=0}^{w-1}[I(i,j)-K(i,j)]^2\\
+        PSNR&=10*log_{10}(\frac{MAX^2}{MSE})
+
+    Args:
+        max_val (Union[int, float]): The dynamic range of the pixel values (255 for 8-bit grayscale images).
+          Default: 1.0.
+
+    Inputs:
+        - **img1** (Tensor) - The first image batch with format 'NCHW'. It should be the same shape and dtype as img2.
+        - **img2** (Tensor) - The second image batch with format 'NCHW'. It should be the same shape and dtype as img1.
+
+    Outputs:
+        Tensor, with dtype mindspore.float32. It is a 1-D tensor with shape N, where N is the batch num of img1.
+
+    Examples:
+        >>> net = nn.PSNR()
+        >>> img1 = Tensor(np.random.random((1,3,16,16)))
+        >>> img2 = Tensor(np.random.random((1,3,16,16)))
+        >>> psnr = net(img1, img2)
+
+    """
+    def __init__(self, max_val=1.0):
+        super(PSNR, self).__init__()
+        validator.check_type('max_val', max_val, [int, float])
+        validator.check('max_val', max_val, '', 0.0, Rel.GT)
+        self.max_val = max_val
+
+    def construct(self, img1, img2):
+        max_val = _convert_img_dtype_to_float32(self.max_val, self.max_val)
+        img1 = _convert_img_dtype_to_float32(img1, self.max_val)
+        img2 = _convert_img_dtype_to_float32(img2, self.max_val)
+
+        mse = P.ReduceMean()(F.square(img1 - img2), (-3, -2, -1))
+        # 10*log_10(max_val^2/MSE)
+        psnr = 10 * P.Log()(F.square(max_val) / mse) / F.scalar_log(10.0)
+
+        return psnr

tests/ut/python/nn/test_psnr.py

+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# 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.
+# ============================================================================
+"""
+test psnr
+"""
+import numpy as np
+import pytest
+import mindspore.nn as nn
+from mindspore.common.api import _executor
+from mindspore import Tensor
+
+
+class PSNRNet(nn.Cell):
+    def __init__(self, max_val=1.0):
+        super(PSNRNet, self).__init__()
+        self.net = nn.PSNR(max_val)
+
+    def construct(self, img1, img2):
+        return self.net(img1, img2)
+
+
+def test_compile_psnr():
+    max_val = 1.0
+    net = PSNRNet(max_val)
+    img1 = Tensor(np.random.random((8, 3, 16, 16)))
+    img2 = Tensor(np.random.random((8, 3, 16, 16)))
+    _executor.compile(net, img1, img2)
+
+def test_compile_psnr_grayscale():
+    max_val = 255
+    net = PSNRNet(max_val)
+    img1 = Tensor(np.random.randint(0, 256, (8, 1, 16, 16), np.uint8))
+    img2 = Tensor(np.random.randint(0, 256, (8, 1, 16, 16), np.uint8))
+    _executor.compile(net, img1, img2)
+
+def test_psnr_max_val_negative():
+    max_val = -1
+    with pytest.raises(ValueError):
+        net = PSNRNet(max_val)
+
+def test_psnr_max_val_bool():
+    max_val = True
+    with pytest.raises(ValueError):
+        net = PSNRNet(max_val)
+
+def test_psnr_max_val_zero():
+    max_val = 0
+    with pytest.raises(ValueError):
+        net = PSNRNet(max_val)

tests/ut/python/nn/test_ssim.py

+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# 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.
+# ============================================================================
+"""
+test ssim
+"""
+import numpy as np
+import pytest
+import mindspore.nn as nn
+from mindspore.common.api import _executor
+from mindspore import Tensor
+
+
+class SSIMNet(nn.Cell):
+    def __init__(self, max_val=1.0, filter_size=11, filter_sigma=1.5, k1=0.01, k2=0.03):
+        super(SSIMNet, self).__init__()
+        self.net = nn.SSIM(max_val, filter_size, filter_sigma, k1, k2)
+
+    def construct(self, img1, img2):
+        return self.net(img1, img2)
+
+
+def test_compile():
+    net = SSIMNet()
+    img1 = Tensor(np.random.random((8, 3, 16, 16)))
+    img2 = Tensor(np.random.random((8, 3, 16, 16)))
+    _executor.compile(net, img1, img2)
+
+def test_compile_grayscale():
+    max_val = 255
+    net = SSIMNet(max_val = max_val)
+    img1 = Tensor(np.random.randint(0, 256, (8, 1, 16, 16), np.uint8))
+    img2 = Tensor(np.random.randint(0, 256, (8, 1, 16, 16), np.uint8))
+    _executor.compile(net, img1, img2)
+
+def test_ssim_max_val_negative():
+    max_val = -1
+    with pytest.raises(ValueError):
+        net = SSIMNet(max_val)
+
+def test_ssim_max_val_bool():
+    max_val = True
+    with pytest.raises(ValueError):
+        net = SSIMNet(max_val)
+
+def test_ssim_max_val_zero():
+    max_val = 0
+    with pytest.raises(ValueError):
+        net = SSIMNet(max_val)
+
+def test_ssim_filter_size_float():
+    with pytest.raises(ValueError):
+        net = SSIMNet(filter_size=1.1)
+
+def test_ssim_filter_size_zero():
+    with pytest.raises(ValueError):
+        net = SSIMNet(filter_size=0)
+
+def test_ssim_filter_sigma_zero():
+    with pytest.raises(ValueError):
+        net = SSIMNet(filter_sigma=0.0)
+
+def test_ssim_filter_sigma_negative():
+    with pytest.raises(ValueError):
+        net = SSIMNet(filter_sigma=-0.1)
+
+def test_ssim_k1_k2_wrong_value():
+    with pytest.raises(ValueError):
+        net = SSIMNet(k1=1.1)
+    with pytest.raises(ValueError):
+        net = SSIMNet(k1=1.0)
+    with pytest.raises(ValueError):
+        net = SSIMNet(k1=0.0)
+    with pytest.raises(ValueError):
+        net = SSIMNet(k1=-1.0)
+
+    with pytest.raises(ValueError):
+        net = SSIMNet(k2=1.1)
+    with pytest.raises(ValueError):
+        net = SSIMNet(k2=1.0)
+    with pytest.raises(ValueError):
+        net = SSIMNet(k2=0.0)
+    with pytest.raises(ValueError):
+        net = SSIMNet(k2=-1.0)
\ No newline at end of file