363-update gaussian 1d (#1020)

wyli · web-flow · commit e6ac8bf3d77e · 2020-09-11T13:21:54.000+01:00
* update gaussian 1d

Signed-off-by: Wenqi Li &lt;wenqil@nvidia.com&gt;

* update gaussian related test cases

Signed-off-by: Wenqi Li &lt;wenqil@nvidia.com&gt;

* [MONAI] python code formatting

Signed-off-by: monai-bot &lt;monai.miccai2019@gmail.com&gt;
diff --git a/monai/networks/layers/convutils.py b/monai/networks/layers/convutils.py
@@ -12,6 +12,7 @@
 from typing import Sequence, Tuple, Union
 
 import numpy as np
+import torch
 
 __all__ = ["same_padding", "stride_minus_kernel_padding", "calculate_out_shape", "gaussian_1d"]
 
@@ -77,7 +78,7 @@ def calculate_out_shape(
     return out_shape if len(out_shape) > 1 else out_shape[0]
 
 
-def gaussian_1d(sigma: float, truncated: float = 4.0) -> np.ndarray:
+def gaussian_1d(sigma: Union[float, torch.Tensor], truncated: float = 4.0) -> np.ndarray:
     """
     one dimensional gaussian kernel.
 
@@ -86,18 +87,17 @@ def gaussian_1d(sigma: float, truncated: float = 4.0) -> np.ndarray:
         truncated: tail length
 
     Raises:
-        ValueError: When ``sigma`` is nonpositive.
+        ValueError: When ``sigma`` is non-positive.
 
     Returns:
-        1D numpy array
+        1D torch tensor
 
     """
-    if sigma <= 0:
-        raise ValueError(f"sigma must be positive, got {sigma}.")
-
+    sigma = torch.as_tensor(sigma).float()
+    if sigma <= 0 or truncated <= 0:
+        raise ValueError(f"sigma and truncated must be positive, got {sigma} and {truncated}.")
     tail = int(sigma * truncated + 0.5)
-    sigma2 = sigma * sigma
-    x = np.arange(-tail, tail + 1)
-    out = np.exp(-0.5 / sigma2 * x ** 2)
-    out /= out.sum()
-    return out
+    x = torch.arange(-tail, tail + 1).float()
+    t = 1 / (torch.tensor(2.0).sqrt() * sigma)
+    out = 0.5 * ((t * (x + 0.5)).erf() - (t * (x - 0.5)).erf())
+    return out.clamp(min=0)
diff --git a/tests/test_gaussian.py b/tests/test_gaussian.py
@@ -0,0 +1,54 @@
+# Copyright 2020 MONAI Consortium
+# 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.
+
+import unittest
+
+import numpy as np
+import torch
+
+from monai.networks.layers.convutils import gaussian_1d
+
+
+class TestGaussian1d(unittest.TestCase):
+    def test_gaussian(self):
+        np.testing.assert_allclose(
+            gaussian_1d(0.5, 8),
+            torch.tensor(
+                [
+                    0.0000e00,
+                    2.9802e-07,
+                    1.3496e-03,
+                    1.5731e-01,
+                    6.8269e-01,
+                    1.5731e-01,
+                    1.3496e-03,
+                    2.9802e-07,
+                    0.0000e00,
+                ]
+            ),
+            rtol=1e-4,
+        )
+
+        np.testing.assert_allclose(
+            gaussian_1d(1, 1),
+            torch.tensor([0.24173, 0.382925, 0.24173]),
+            rtol=1e-4,
+        )
+
+    def test_wrong_sigma(self):
+        with self.assertRaises(ValueError):
+            gaussian_1d(-1, 10)
+        with self.assertRaises(ValueError):
+            gaussian_1d(1, -10)
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_gaussian_filter.py b/tests/test_gaussian_filter.py
@@ -25,16 +25,16 @@ def test_1d(self):
             [
                 [
                     [
-                        0.56658804,
-                        0.69108766,
-                        0.79392236,
-                        0.86594427,
-                        0.90267116,
-                        0.9026711,
-                        0.8659443,
-                        0.7939224,
-                        0.6910876,
-                        0.56658804,
+                        0.5654129,
+                        0.68915915,
+                        0.79146194,
+                        0.8631974,
+                        0.8998163,
+                        0.8998163,
+                        0.8631973,
+                        0.79146194,
+                        0.6891592,
+                        0.5654129,
                     ]
                 ]
             ]
@@ -49,9 +49,9 @@ def test_2d(self):
             [
                 [
                     [
-                        [0.13380532, 0.14087981, 0.13380532],
-                        [0.14087981, 0.14832835, 0.14087981],
-                        [0.13380532, 0.14087981, 0.13380532],
+                        [0.13239081, 0.13932934, 0.13239081],
+                        [0.13932936, 0.14663152, 0.13932936],
+                        [0.13239081, 0.13932934, 0.13239081],
                     ]
                 ]
             ]
@@ -65,29 +65,30 @@ def test_2d(self):
     def test_3d(self):
         a = torch.ones(1, 1, 4, 3, 4)
         g = GaussianFilter(3, 3, 3).to(torch.device("cpu:0"))
+
         expected = np.array(
             [
                 [
                     [
                         [
-                            [0.07294822, 0.08033235, 0.08033235, 0.07294822],
-                            [0.07680509, 0.08457965, 0.08457965, 0.07680509],
-                            [0.07294822, 0.08033235, 0.08033235, 0.07294822],
+                            [0.07189433, 0.07911152, 0.07911152, 0.07189433],
+                            [0.07566228, 0.08325771, 0.08325771, 0.07566228],
+                            [0.07189433, 0.07911152, 0.07911152, 0.07189433],
                         ],
                         [
-                            [0.08033235, 0.08846395, 0.08846395, 0.08033235],
-                            [0.08457965, 0.09314119, 0.09314119, 0.08457966],
-                            [0.08033235, 0.08846396, 0.08846396, 0.08033236],
+                            [0.07911152, 0.08705322, 0.08705322, 0.07911152],
+                            [0.08325771, 0.09161563, 0.09161563, 0.08325771],
+                            [0.07911152, 0.08705322, 0.08705322, 0.07911152],
                         ],
                         [
-                            [0.08033235, 0.08846395, 0.08846395, 0.08033235],
-                            [0.08457965, 0.09314119, 0.09314119, 0.08457966],
-                            [0.08033235, 0.08846396, 0.08846396, 0.08033236],
+                            [0.07911152, 0.08705322, 0.08705322, 0.07911152],
+                            [0.08325771, 0.09161563, 0.09161563, 0.08325771],
+                            [0.07911152, 0.08705322, 0.08705322, 0.07911152],
                         ],
                         [
-                            [0.07294822, 0.08033235, 0.08033235, 0.07294822],
-                            [0.07680509, 0.08457965, 0.08457965, 0.07680509],
-                            [0.07294822, 0.08033235, 0.08033235, 0.07294822],
+                            [0.07189433, 0.07911152, 0.07911152, 0.07189433],
+                            [0.07566228, 0.08325771, 0.08325771, 0.07566228],
+                            [0.07189433, 0.07911152, 0.07911152, 0.07189433],
                         ],
                     ]
                 ]
@@ -98,14 +99,15 @@ def test_3d(self):
     def test_3d_sigmas(self):
         a = torch.ones(1, 1, 4, 3, 2)
         g = GaussianFilter(3, [3, 2, 1], 3).to(torch.device("cpu:0"))
+
         expected = np.array(
             [
                 [
                     [
-                        [[0.1422854, 0.1422854], [0.15806103, 0.15806103], [0.1422854, 0.1422854]],
-                        [[0.15668818, 0.15668817], [0.17406069, 0.17406069], [0.15668818, 0.15668817]],
-                        [[0.15668818, 0.15668817], [0.17406069, 0.17406069], [0.15668818, 0.15668817]],
-                        [[0.1422854, 0.1422854], [0.15806103, 0.15806103], [0.1422854, 0.1422854]],
+                        [[0.13690521, 0.13690521], [0.15181276, 0.15181276], [0.13690521, 0.13690521]],
+                        [[0.1506486, 0.15064861], [0.16705267, 0.16705267], [0.1506486, 0.15064861]],
+                        [[0.1506486, 0.15064861], [0.16705267, 0.16705267], [0.1506486, 0.15064861]],
+                        [[0.13690521, 0.13690521], [0.15181276, 0.15181276], [0.13690521, 0.13690521]],
                     ]
                 ]
             ]
diff --git a/tests/test_gaussian_sharpen.py b/tests/test_gaussian_sharpen.py
@@ -21,8 +21,8 @@
     np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),
     np.array(
         [
-            [[4.0335875, 3.362756, 4.0335875], [3.588128, 2.628216, 3.588128], [4.491922, 3.8134987, 4.491922]],
-            [[10.427719, 8.744948, 10.427719], [8.97032, 6.5705404, 8.970321], [10.886056, 9.195692, 10.886056]],
+            [[4.1081963, 3.4950666, 4.1081963], [3.7239995, 2.8491793, 3.7239995], [4.569839, 3.9529324, 4.569839]],
+            [[10.616725, 9.081067, 10.616725], [9.309998, 7.12295, 9.309998], [11.078365, 9.538931, 11.078365]],
         ]
     ),
 ]
@@ -32,8 +32,8 @@
     np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),
     np.array(
         [
-            [[4.146659, 4.392873, 4.146659], [8.031006, 8.804623, 8.031005], [10.127394, 11.669131, 10.127394]],
-            [[14.852196, 16.439377, 14.852201], [20.077503, 22.011555, 20.077507], [20.832941, 23.715641, 20.832935]],
+            [[4.513644, 4.869134, 4.513644], [8.467242, 9.4004135, 8.467242], [10.416813, 12.0653515, 10.416813]],
+            [[15.711488, 17.569994, 15.711488], [21.16811, 23.501041, 21.16811], [21.614658, 24.766209, 21.614658]],
         ]
     ),
 ]
@@ -43,8 +43,8 @@
     np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),
     np.array(
         [
-            [[3.129089, 3.0711129, 3.129089], [6.783306, 6.8526435, 6.7833037], [11.901203, 13.098082, 11.901203]],
-            [[14.401806, 15.198004, 14.401809], [16.958261, 17.131605, 16.958261], [23.17392, 25.224974, 23.17392]],
+            [[3.3324685, 3.335536, 3.3324673], [7.7666636, 8.16056, 7.7666636], [12.662973, 14.317837, 12.6629715]],
+            [[15.329051, 16.57557, 15.329051], [19.41665, 20.40139, 19.416655], [24.659554, 27.557873, 24.659554]],
         ]
     ),
 ]
diff --git a/tests/test_gaussian_sharpend.py b/tests/test_gaussian_sharpend.py
@@ -21,8 +21,8 @@
     {"img": np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]])},
     np.array(
         [
-            [[4.0335875, 3.362756, 4.0335875], [3.588128, 2.628216, 3.588128], [4.491922, 3.8134987, 4.491922]],
-            [[10.427719, 8.744948, 10.427719], [8.97032, 6.5705404, 8.970321], [10.886056, 9.195692, 10.886056]],
+            [[4.1081963, 3.4950666, 4.1081963], [3.7239995, 2.8491793, 3.7239995], [4.569839, 3.9529324, 4.569839]],
+            [[10.616725, 9.081067, 10.616725], [9.309998, 7.12295, 9.309998], [11.078365, 9.538931, 11.078365]],
         ]
     ),
 ]
@@ -32,8 +32,8 @@
     {"img": np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]])},
     np.array(
         [
-            [[4.146659, 4.392873, 4.146659], [8.031006, 8.804623, 8.031005], [10.127394, 11.669131, 10.127394]],
-            [[14.852196, 16.439377, 14.852201], [20.077503, 22.011555, 20.077507], [20.832941, 23.715641, 20.832935]],
+            [[4.513644, 4.869134, 4.513644], [8.467242, 9.4004135, 8.467242], [10.416813, 12.0653515, 10.416813]],
+            [[15.711488, 17.569994, 15.711488], [21.16811, 23.501041, 21.16811], [21.614658, 24.766209, 21.614658]],
         ]
     ),
 ]
@@ -43,8 +43,8 @@
     {"img": np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]])},
     np.array(
         [
-            [[3.129089, 3.0711129, 3.129089], [6.783306, 6.8526435, 6.7833037], [11.901203, 13.098082, 11.901203]],
-            [[14.401806, 15.198004, 14.401809], [16.958261, 17.131605, 16.958261], [23.17392, 25.224974, 23.17392]],
+            [[3.3324685, 3.335536, 3.3324673], [7.7666636, 8.16056, 7.7666636], [12.662973, 14.317837, 12.6629715]],
+            [[15.329051, 16.57557, 15.329051], [19.41665, 20.40139, 19.416655], [24.659554, 27.557873, 24.659554]],
         ]
     ),
 ]
diff --git a/tests/test_gaussian_smooth.py b/tests/test_gaussian_smooth.py
@@ -21,8 +21,12 @@
     np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),
     np.array(
         [
-            [[0.5999930, 0.7056839, 0.5999930], [0.8140513, 0.9574494, 0.8140513], [0.7842673, 0.9224188, 0.7842673]],
-            [[1.6381884, 1.926761, 1.6381884], [2.0351284, 2.3936234, 2.0351284], [1.8224627, 2.143496, 1.8224627]],
+            [
+                [0.59167546, 0.69312394, 0.59167546],
+                [0.7956997, 0.93213004, 0.7956997],
+                [0.7668002, 0.8982755, 0.7668002],
+            ],
+            [[1.6105323, 1.8866735, 1.6105323], [1.9892492, 2.3303251, 1.9892492], [1.7856569, 2.091825, 1.7856569]],
         ]
     ),
 ]
@@ -32,8 +36,8 @@
     np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),
     np.array(
         [
-            [[0.893521, 0.99973595, 0.893521], [1.785628, 1.9978896, 1.7856278], [2.2983139, 2.5715199, 2.2983139]],
-            [[3.2873974, 3.6781778, 3.2873974], [4.46407, 4.9947243, 4.46407], [4.69219, 5.2499614, 4.69219]],
+            [[0.8424794, 0.99864554, 0.8424794], [1.678146, 1.9892154, 1.678146], [1.9889624, 2.3576462, 1.9889624]],
+            [[2.966061, 3.5158648, 2.966061], [4.1953645, 4.973038, 4.1953645], [4.112544, 4.8748655, 4.1125436]],
         ]
     ),
 ]
@@ -43,8 +47,8 @@
     np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),
     np.array(
         [
-            [[0.91108215, 1.0193846, 0.91108215], [1.236127, 1.3830683, 1.236127], [1.1909003, 1.3324654, 1.1909003]],
-            [[2.4875693, 2.7832723, 2.487569], [3.0903177, 3.457671, 3.0903175], [2.7673876, 3.0963533, 2.7673874]],
+            [[0.8542037, 1.0125432, 0.8542037], [1.1487541, 1.3616928, 1.1487541], [1.1070318, 1.3122368, 1.1070318]],
+            [[2.3251305, 2.756128, 2.3251305], [2.8718853, 3.4042323, 2.8718853], [2.5779586, 3.0558217, 2.5779586]],
         ]
     ),
 ]
diff --git a/tests/test_gaussian_smoothd.py b/tests/test_gaussian_smoothd.py
@@ -21,8 +21,12 @@
     {"img": np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]])},
     np.array(
         [
-            [[0.5999930, 0.7056839, 0.5999930], [0.8140513, 0.9574494, 0.8140513], [0.7842673, 0.9224188, 0.7842673]],
-            [[1.6381884, 1.926761, 1.6381884], [2.0351284, 2.3936234, 2.0351284], [1.8224627, 2.143496, 1.8224627]],
+            [
+                [0.59167546, 0.69312394, 0.59167546],
+                [0.7956997, 0.93213004, 0.7956997],
+                [0.7668002, 0.8982755, 0.7668002],
+            ],
+            [[1.6105323, 1.8866735, 1.6105323], [1.9892492, 2.3303251, 1.9892492], [1.7856569, 2.091825, 1.7856569]],
         ]
     ),
 ]
@@ -32,8 +36,8 @@
     {"img": np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]])},
     np.array(
         [
-            [[0.893521, 0.99973595, 0.893521], [1.785628, 1.9978896, 1.7856278], [2.2983139, 2.5715199, 2.2983139]],
-            [[3.2873974, 3.6781778, 3.2873974], [4.46407, 4.9947243, 4.46407], [4.69219, 5.2499614, 4.69219]],
+            [[0.8424794, 0.99864554, 0.8424794], [1.678146, 1.9892154, 1.678146], [1.9889624, 2.3576462, 1.9889624]],
+            [[2.966061, 3.5158648, 2.966061], [4.1953645, 4.973038, 4.1953645], [4.112544, 4.8748655, 4.1125436]],
         ]
     ),
 ]
@@ -43,8 +47,8 @@
     {"img": np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]])},
     np.array(
         [
-            [[0.91108215, 1.0193846, 0.91108215], [1.236127, 1.3830683, 1.236127], [1.1909003, 1.3324654, 1.1909003]],
-            [[2.4875693, 2.7832723, 2.487569], [3.0903177, 3.457671, 3.0903175], [2.7673876, 3.0963533, 2.7673874]],
+            [[0.8542037, 1.0125432, 0.8542037], [1.1487541, 1.3616928, 1.1487541], [1.1070318, 1.3122368, 1.1070318]],
+            [[2.3251305, 2.756128, 2.3251305], [2.8718853, 3.4042323, 2.8718853], [2.5779586, 3.0558217, 2.5779586]],
         ]
     ),
 ]
diff --git a/tests/test_rand_elastic_3d.py b/tests/test_rand_elastic_3d.py
@@ -50,7 +50,7 @@
             "device": None,
         },
         {"img": torch.arange(27).reshape((1, 3, 3, 3)), "spatial_size": (2, 2, 2)},
-        np.array([[[[6.492354, 7.5022864], [9.519528, 10.524366]], [[15.51277, 16.525297], [18.533852, 19.539217]]]]),
+        np.array([[[[6.4939356, 7.50289], [9.518351, 10.522849]], [[15.512375, 16.523542], [18.531467, 19.53646]]]]),
     ],
     [
         {
@@ -63,7 +63,7 @@
             "spatial_size": (2, 2, 2),
         },
         {"img": torch.arange(27).reshape((1, 3, 3, 3)), "mode": "bilinear"},
-        np.array([[[[5.005563, 9.463698], [9.289501, 13.741863]], [[12.320587, 16.779654], [16.597677, 21.049414]]]]),
+        np.array([[[[5.0069294, 9.463932], [9.287769, 13.739735]], [[12.319424, 16.777205], [16.594296, 21.045748]]]]),
     ],
 ]
 
diff --git a/tests/test_rand_elasticd_3d.py b/tests/test_rand_elasticd_3d.py
@@ -68,7 +68,7 @@
             "spatial_size": (2, 2, 2),
         },
         {"img": torch.arange(27).reshape((1, 3, 3, 3)), "seg": torch.arange(27).reshape((1, 3, 3, 3))},
-        np.array([[[[6.492354, 7.5022864], [9.519528, 10.524366]], [[15.51277, 16.525297], [18.533852, 19.539217]]]]),
+        np.array([[[[6.4939356, 7.50289], [9.518351, 10.522849]], [[15.512375, 16.523542], [18.531467, 19.53646]]]]),
     ],
     [
         {
@@ -83,7 +83,7 @@
             "mode": "bilinear",
         },
         {"img": torch.arange(27).reshape((1, 3, 3, 3)), "seg": torch.arange(27).reshape((1, 3, 3, 3))},
-        np.array([[[[5.005563, 9.463698], [9.289501, 13.741863]], [[12.320587, 16.779654], [16.597677, 21.049414]]]]),
+        np.array([[[[5.0069294, 9.463932], [9.287769, 13.739735]], [[12.319424, 16.777205], [16.594296, 21.045748]]]]),
     ],
     [
         {
@@ -99,7 +99,7 @@
         },
         {"img": torch.arange(27).reshape((1, 3, 3, 3)), "seg": torch.arange(27).reshape((1, 3, 3, 3))},
         {
-            "img": torch.tensor([[[[5.0056, 9.4637], [9.2895, 13.7419]], [[12.3206, 16.7797], [16.5977, 21.0494]]]]),
+            "img": torch.tensor([[[[5.0069, 9.4639], [9.2878, 13.7397]], [[12.3194, 16.7772], [16.5943, 21.0457]]]]),
             "seg": torch.tensor([[[[4.0, 14.0], [7.0, 14.0]], [[9.0, 19.0], [12.0, 22.0]]]]),
         },
     ],
diff --git a/tests/test_rand_gaussian_sharpen.py b/tests/test_rand_gaussian_sharpen.py
diff --git a/tests/test_rand_gaussian_sharpend.py b/tests/test_rand_gaussian_sharpend.py
diff --git a/tests/test_rand_gaussian_smooth.py b/tests/test_rand_gaussian_smooth.py
diff --git a/tests/test_rand_gaussian_smoothd.py b/tests/test_rand_gaussian_smoothd.py

Original file line number	Diff line number	Diff line change
`@@ -25,16 +25,16 @@ def test_1d(self):`
`25`	`25`	`[`
`26`	`26`	`[`
`27`	`27`	`[`
`28`		`- 0.56658804,`
`29`		`- 0.69108766,`
`30`		`- 0.79392236,`
`31`		`- 0.86594427,`
`32`		`- 0.90267116,`
`33`		`- 0.9026711,`
`34`		`- 0.8659443,`
`35`		`- 0.7939224,`
`36`		`- 0.6910876,`
`37`		`- 0.56658804,`
	`28`	`+ 0.5654129,`
	`29`	`+ 0.68915915,`
	`30`	`+ 0.79146194,`
	`31`	`+ 0.8631974,`
	`32`	`+ 0.8998163,`
	`33`	`+ 0.8998163,`
	`34`	`+ 0.8631973,`
	`35`	`+ 0.79146194,`
	`36`	`+ 0.6891592,`
	`37`	`+ 0.5654129,`
`38`	`38`	`]`
`39`	`39`	`]`
`40`	`40`	`]`
`@@ -49,9 +49,9 @@ def test_2d(self):`
`49`	`49`	`[`
`50`	`50`	`[`
`51`	`51`	`[`
`52`		`- [0.13380532, 0.14087981, 0.13380532],`
`53`		`- [0.14087981, 0.14832835, 0.14087981],`
`54`		`- [0.13380532, 0.14087981, 0.13380532],`
	`52`	`+ [0.13239081, 0.13932934, 0.13239081],`
	`53`	`+ [0.13932936, 0.14663152, 0.13932936],`
	`54`	`+ [0.13239081, 0.13932934, 0.13239081],`
`55`	`55`	`]`
`56`	`56`	`]`
`57`	`57`	`]`
`@@ -65,29 +65,30 @@ def test_2d(self):`
`65`	`65`	`def test_3d(self):`
`66`	`66`	`a = torch.ones(1, 1, 4, 3, 4)`
`67`	`67`	`g = GaussianFilter(3, 3, 3).to(torch.device("cpu:0"))`
	`68`	`+`
`68`	`69`	`expected = np.array(`
`69`	`70`	`[`
`70`	`71`	`[`
`71`	`72`	`[`
`72`	`73`	`[`
`73`		`- [0.07294822, 0.08033235, 0.08033235, 0.07294822],`
`74`		`- [0.07680509, 0.08457965, 0.08457965, 0.07680509],`
`75`		`- [0.07294822, 0.08033235, 0.08033235, 0.07294822],`
	`74`	`+ [0.07189433, 0.07911152, 0.07911152, 0.07189433],`
	`75`	`+ [0.07566228, 0.08325771, 0.08325771, 0.07566228],`
	`76`	`+ [0.07189433, 0.07911152, 0.07911152, 0.07189433],`
`76`	`77`	`],`
`77`	`78`	`[`
`78`		`- [0.08033235, 0.08846395, 0.08846395, 0.08033235],`
`79`		`- [0.08457965, 0.09314119, 0.09314119, 0.08457966],`
`80`		`- [0.08033235, 0.08846396, 0.08846396, 0.08033236],`
	`79`	`+ [0.07911152, 0.08705322, 0.08705322, 0.07911152],`
	`80`	`+ [0.08325771, 0.09161563, 0.09161563, 0.08325771],`
	`81`	`+ [0.07911152, 0.08705322, 0.08705322, 0.07911152],`
`81`	`82`	`],`
`82`	`83`	`[`
`83`		`- [0.08033235, 0.08846395, 0.08846395, 0.08033235],`
`84`		`- [0.08457965, 0.09314119, 0.09314119, 0.08457966],`
`85`		`- [0.08033235, 0.08846396, 0.08846396, 0.08033236],`
	`84`	`+ [0.07911152, 0.08705322, 0.08705322, 0.07911152],`
	`85`	`+ [0.08325771, 0.09161563, 0.09161563, 0.08325771],`
	`86`	`+ [0.07911152, 0.08705322, 0.08705322, 0.07911152],`
`86`	`87`	`],`
`87`	`88`	`[`
`88`		`- [0.07294822, 0.08033235, 0.08033235, 0.07294822],`
`89`		`- [0.07680509, 0.08457965, 0.08457965, 0.07680509],`
`90`		`- [0.07294822, 0.08033235, 0.08033235, 0.07294822],`
	`89`	`+ [0.07189433, 0.07911152, 0.07911152, 0.07189433],`
	`90`	`+ [0.07566228, 0.08325771, 0.08325771, 0.07566228],`
	`91`	`+ [0.07189433, 0.07911152, 0.07911152, 0.07189433],`
`91`	`92`	`],`
`92`	`93`	`]`
`93`	`94`	`]`
`@@ -98,14 +99,15 @@ def test_3d(self):`
`98`	`99`	`def test_3d_sigmas(self):`
`99`	`100`	`a = torch.ones(1, 1, 4, 3, 2)`
`100`	`101`	`g = GaussianFilter(3, [3, 2, 1], 3).to(torch.device("cpu:0"))`
	`102`	`+`
`101`	`103`	`expected = np.array(`
`102`	`104`	`[`
`103`	`105`	`[`
`104`	`106`	`[`
`105`		`- [[0.1422854, 0.1422854], [0.15806103, 0.15806103], [0.1422854, 0.1422854]],`
`106`		`- [[0.15668818, 0.15668817], [0.17406069, 0.17406069], [0.15668818, 0.15668817]],`
`107`		`- [[0.15668818, 0.15668817], [0.17406069, 0.17406069], [0.15668818, 0.15668817]],`
`108`		`- [[0.1422854, 0.1422854], [0.15806103, 0.15806103], [0.1422854, 0.1422854]],`
	`107`	`+ [[0.13690521, 0.13690521], [0.15181276, 0.15181276], [0.13690521, 0.13690521]],`
	`108`	`+ [[0.1506486, 0.15064861], [0.16705267, 0.16705267], [0.1506486, 0.15064861]],`
	`109`	`+ [[0.1506486, 0.15064861], [0.16705267, 0.16705267], [0.1506486, 0.15064861]],`
	`110`	`+ [[0.13690521, 0.13690521], [0.15181276, 0.15181276], [0.13690521, 0.13690521]],`
`109`	`111`	`]`
`110`	`112`	`]`
`111`	`113`	`]`
Original file line number	Diff line number	Diff line change
`@@ -21,8 +21,8 @@`
`21`	`21`	`np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),`
`22`	`22`	`np.array(`
`23`	`23`	`[`
`24`		`- [[4.0335875, 3.362756, 4.0335875], [3.588128, 2.628216, 3.588128], [4.491922, 3.8134987, 4.491922]],`
`25`		`- [[10.427719, 8.744948, 10.427719], [8.97032, 6.5705404, 8.970321], [10.886056, 9.195692, 10.886056]],`
	`24`	`+ [[4.1081963, 3.4950666, 4.1081963], [3.7239995, 2.8491793, 3.7239995], [4.569839, 3.9529324, 4.569839]],`
	`25`	`+ [[10.616725, 9.081067, 10.616725], [9.309998, 7.12295, 9.309998], [11.078365, 9.538931, 11.078365]],`
`26`	`26`	`]`
`27`	`27`	`),`
`28`	`28`	`]`
`@@ -32,8 +32,8 @@`
`32`	`32`	`np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),`
`33`	`33`	`np.array(`
`34`	`34`	`[`
`35`		`- [[4.146659, 4.392873, 4.146659], [8.031006, 8.804623, 8.031005], [10.127394, 11.669131, 10.127394]],`
`36`		`- [[14.852196, 16.439377, 14.852201], [20.077503, 22.011555, 20.077507], [20.832941, 23.715641, 20.832935]],`
	`35`	`+ [[4.513644, 4.869134, 4.513644], [8.467242, 9.4004135, 8.467242], [10.416813, 12.0653515, 10.416813]],`
	`36`	`+ [[15.711488, 17.569994, 15.711488], [21.16811, 23.501041, 21.16811], [21.614658, 24.766209, 21.614658]],`
`37`	`37`	`]`
`38`	`38`	`),`
`39`	`39`	`]`
`@@ -43,8 +43,8 @@`
`43`	`43`	`np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),`
`44`	`44`	`np.array(`
`45`	`45`	`[`
`46`		`- [[3.129089, 3.0711129, 3.129089], [6.783306, 6.8526435, 6.7833037], [11.901203, 13.098082, 11.901203]],`
`47`		`- [[14.401806, 15.198004, 14.401809], [16.958261, 17.131605, 16.958261], [23.17392, 25.224974, 23.17392]],`
	`46`	`+ [[3.3324685, 3.335536, 3.3324673], [7.7666636, 8.16056, 7.7666636], [12.662973, 14.317837, 12.6629715]],`
	`47`	`+ [[15.329051, 16.57557, 15.329051], [19.41665, 20.40139, 19.416655], [24.659554, 27.557873, 24.659554]],`
`48`	`48`	`]`
`49`	`49`	`),`
`50`	`50`	`]`
Original file line number	Diff line number	Diff line change
`@@ -21,8 +21,12 @@`
`21`	`21`	`np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),`
`22`	`22`	`np.array(`
`23`	`23`	`[`
`24`		`- [[0.5999930, 0.7056839, 0.5999930], [0.8140513, 0.9574494, 0.8140513], [0.7842673, 0.9224188, 0.7842673]],`
`25`		`- [[1.6381884, 1.926761, 1.6381884], [2.0351284, 2.3936234, 2.0351284], [1.8224627, 2.143496, 1.8224627]],`
	`24`	`+ [`
	`25`	`+ [0.59167546, 0.69312394, 0.59167546],`
	`26`	`+ [0.7956997, 0.93213004, 0.7956997],`
	`27`	`+ [0.7668002, 0.8982755, 0.7668002],`
	`28`	`+ ],`
	`29`	`+ [[1.6105323, 1.8866735, 1.6105323], [1.9892492, 2.3303251, 1.9892492], [1.7856569, 2.091825, 1.7856569]],`
`26`	`30`	`]`
`27`	`31`	`),`
`28`	`32`	`]`
`@@ -32,8 +36,8 @@`
`32`	`36`	`np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),`
`33`	`37`	`np.array(`
`34`	`38`	`[`
`35`		`- [[0.893521, 0.99973595, 0.893521], [1.785628, 1.9978896, 1.7856278], [2.2983139, 2.5715199, 2.2983139]],`
`36`		`- [[3.2873974, 3.6781778, 3.2873974], [4.46407, 4.9947243, 4.46407], [4.69219, 5.2499614, 4.69219]],`
	`39`	`+ [[0.8424794, 0.99864554, 0.8424794], [1.678146, 1.9892154, 1.678146], [1.9889624, 2.3576462, 1.9889624]],`
	`40`	`+ [[2.966061, 3.5158648, 2.966061], [4.1953645, 4.973038, 4.1953645], [4.112544, 4.8748655, 4.1125436]],`
`37`	`41`	`]`
`38`	`42`	`),`
`39`	`43`	`]`
`@@ -43,8 +47,8 @@`
`43`	`47`	`np.array([[[1, 1, 1], [2, 2, 2], [3, 3, 3]], [[4, 4, 4], [5, 5, 5], [6, 6, 6]]]),`
`44`	`48`	`np.array(`
`45`	`49`	`[`
`46`		`- [[0.91108215, 1.0193846, 0.91108215], [1.236127, 1.3830683, 1.236127], [1.1909003, 1.3324654, 1.1909003]],`
`47`		`- [[2.4875693, 2.7832723, 2.487569], [3.0903177, 3.457671, 3.0903175], [2.7673876, 3.0963533, 2.7673874]],`
	`50`	`+ [[0.8542037, 1.0125432, 0.8542037], [1.1487541, 1.3616928, 1.1487541], [1.1070318, 1.3122368, 1.1070318]],`
	`51`	`+ [[2.3251305, 2.756128, 2.3251305], [2.8718853, 3.4042323, 2.8718853], [2.5779586, 3.0558217, 2.5779586]],`
`48`	`52`	`]`
`49`	`53`	`),`
`50`	`54`	`]`
Original file line number	Diff line number	Diff line change
`@@ -50,7 +50,7 @@`
`50`	`50`	`"device": None,`
`51`	`51`	`},`
`52`	`52`	`{"img": torch.arange(27).reshape((1, 3, 3, 3)), "spatial_size": (2, 2, 2)},`
`53`		`- np.array([[[[6.492354, 7.5022864], [9.519528, 10.524366]], [[15.51277, 16.525297], [18.533852, 19.539217]]]]),`
	`53`	`+ np.array([[[[6.4939356, 7.50289], [9.518351, 10.522849]], [[15.512375, 16.523542], [18.531467, 19.53646]]]]),`
`54`	`54`	`],`
`55`	`55`	`[`
`56`	`56`	`{`
`@@ -63,7 +63,7 @@`
`63`	`63`	`"spatial_size": (2, 2, 2),`
`64`	`64`	`},`
`65`	`65`	`{"img": torch.arange(27).reshape((1, 3, 3, 3)), "mode": "bilinear"},`
`66`		`- np.array([[[[5.005563, 9.463698], [9.289501, 13.741863]], [[12.320587, 16.779654], [16.597677, 21.049414]]]]),`
	`66`	`+ np.array([[[[5.0069294, 9.463932], [9.287769, 13.739735]], [[12.319424, 16.777205], [16.594296, 21.045748]]]]),`
`67`	`67`	`],`
`68`	`68`	`]`
`69`	`69`
Original file line number	Diff line number	Diff line change
`@@ -68,7 +68,7 @@`
`68`	`68`	`"spatial_size": (2, 2, 2),`
`69`	`69`	`},`
`70`	`70`	`{"img": torch.arange(27).reshape((1, 3, 3, 3)), "seg": torch.arange(27).reshape((1, 3, 3, 3))},`
`71`		`- np.array([[[[6.492354, 7.5022864], [9.519528, 10.524366]], [[15.51277, 16.525297], [18.533852, 19.539217]]]]),`
	`71`	`+ np.array([[[[6.4939356, 7.50289], [9.518351, 10.522849]], [[15.512375, 16.523542], [18.531467, 19.53646]]]]),`
`72`	`72`	`],`
`73`	`73`	`[`
`74`	`74`	`{`
`@@ -83,7 +83,7 @@`
`83`	`83`	`"mode": "bilinear",`
`84`	`84`	`},`
`85`	`85`	`{"img": torch.arange(27).reshape((1, 3, 3, 3)), "seg": torch.arange(27).reshape((1, 3, 3, 3))},`
`86`		`- np.array([[[[5.005563, 9.463698], [9.289501, 13.741863]], [[12.320587, 16.779654], [16.597677, 21.049414]]]]),`
	`86`	`+ np.array([[[[5.0069294, 9.463932], [9.287769, 13.739735]], [[12.319424, 16.777205], [16.594296, 21.045748]]]]),`
`87`	`87`	`],`
`88`	`88`	`[`
`89`	`89`	`{`
`@@ -99,7 +99,7 @@`
`99`	`99`	`},`
`100`	`100`	`{"img": torch.arange(27).reshape((1, 3, 3, 3)), "seg": torch.arange(27).reshape((1, 3, 3, 3))},`
`101`	`101`	`{`
`102`		`- "img": torch.tensor([[[[5.0056, 9.4637], [9.2895, 13.7419]], [[12.3206, 16.7797], [16.5977, 21.0494]]]]),`
	`102`	`+ "img": torch.tensor([[[[5.0069, 9.4639], [9.2878, 13.7397]], [[12.3194, 16.7772], [16.5943, 21.0457]]]]),`
`103`	`103`	`"seg": torch.tensor([[[[4.0, 14.0], [7.0, 14.0]], [[9.0, 19.0], [12.0, 22.0]]]]),`
`104`	`104`	`},`
`105`	`105`	`],`