transpose_and_minmaxnormalize (#288)

* transpose_and_minmaxnormalize

transpose_and_minmaxnormalize

transpose_and_minmaxnormalize

transpose_and_minmaxnormalize

* fix linting

tensorpack/dataflow/imgaug/imgproc.py

@@ -7,7 +7,7 @@ import numpy as np
 import cv2
 
 __all__ = ['Hue', 'Brightness', 'Contrast', 'MeanVarianceNormalize',
-           'GaussianBlur', 'Gamma', 'Clip', 'Saturation', 'Lighting']
+           'GaussianBlur', 'Gamma', 'Clip', 'Saturation', 'Lighting', 'MinMaxNormalize']
 
 
 class Hue(ImageAugmentor):
@@ -98,7 +98,7 @@ class MeanVarianceNormalize(ImageAugmentor):
         Args:
             all_channel (bool): if True, normalize all channels together. else separately.
         """
-        self.all_channel = all_channel
+        self._init(locals())
 
     def _augment(self, img, _):
         img = img.astype('float32')
@@ -232,3 +232,30 @@ class Lighting(ImageAugmentor):
         if old_dtype == np.uint8:
             img = np.clip(img, 0, 255)
         return img.astype(old_dtype)
+
+
+class MinMaxNormalize(ImageAugmentor):
+    """
+    Linearly scales the image to the range [min, max].
+
+    This augmentor always returns float32 images.
+    """
+    def __init__(self, min=0, max=255, all_channel=True):
+        """
+        Args:
+            max (float): The new maximum value
+            min (float): The new minimum value
+            all_channel (bool): if True, normalize all channels together. else separately.
+        """
+        self._init(locals())
+
+    def _augment(self, img, _):
+        img = img.astype('float32')
+        if self.all_channel:
+            minimum = np.min(img)
+            maximum = np.max(img)
+        else:
+            minimum = np.min(img, axis=(0, 1), keepdims=True)
+            maximum = np.max(img, axis=(0, 1), keepdims=True)
+        img = (self.max - self.min) * (img - minimum) / (maximum - minimum) + self.min
+        return img

tensorpack/dataflow/imgaug/noname.py

@@ -8,7 +8,7 @@ from ...utils.argtools import shape2d
 import numpy as np
 import cv2
 
-__all__ = ['Flip', 'Resize', 'RandomResize', 'ResizeShortestEdge']
+__all__ = ['Flip', 'Resize', 'RandomResize', 'ResizeShortestEdge', 'Transpose']
 
 
 class Flip(ImageAugmentor):
@@ -140,3 +140,31 @@ class RandomResize(ImageAugmentor):
         if img.ndim == 3 and ret.ndim == 2:
             ret = ret[:, :, np.newaxis]
         return ret
+
+
+class Transpose(ImageAugmentor):
+    """
+    Random transpose the image
+    """
+    def __init__(self, prob=0.5):
+        """
+        Args:
+            prob (float): probability of transpose.
+        """
+        super(Transpose, self).__init__()
+        self.prob = prob
+        self._init()
+
+    def _get_augment_params(self, img):
+        return self._rand_range() < self.prob
+
+    def _augment(self, img, do):
+        ret = img
+        if do:
+            ret = cv2.transpose(img)
+            if img.ndim == 3 and ret.ndim == 2:
+                ret = ret[:, :, np.newaxis]
+        return ret
+
+    def _fprop_coord(self, coord, param):
+        raise NotImplementedError()