svhn mean

tensorpack/dataflow/dataset/svhn.py

@@ -20,10 +20,15 @@ class SVHNDigit(DataFlow):
     SVHN Cropped Digit Dataset
     return img of 32x32x3, label of 0-9
     """
+    Cache = {}
+
     def __init__(self, name, data_dir=None):
         """
         name: 'train', 'test', or 'extra'
         """
+        if name in SVHNDigit.Cache:
+            self.X, self.Y = SVHNDigit.Cache[name]
+            return
         if data_dir is None:
             data_dir = os.path.join(
                 os.path.dirname(__file__),
@@ -38,6 +43,7 @@ http://ufldl.stanford.edu/housenumbers/".format(filename)
         self.X = data['X'].transpose(3,0,1,2)
         self.Y = data['y'].reshape((-1))
         self.Y[self.Y==10] = 0
+        SVHNDigit.Cache[name] = (self.X, self.Y)
 
     def size(self):
         return self.X.shape[0]
@@ -47,3 +53,16 @@ http://ufldl.stanford.edu/housenumbers/".format(filename)
         for k in range(n):
             yield [self.X[k], self.Y[k]]
 
+    @staticmethod
+    def get_per_pixel_mean():
+        """
+        return 32x32x3 image
+        """
+        a = SVHNDigit('train')
+        b = SVHNDigit('test')
+        c = SVHNDigit('extra')
+        return numpy.concatenate((a.X, b.X, c.X)).mean(axis=0)
+
+if __name__ == '__main__':
+    a = SVHNDigit('train')
+    b = SVHNDigit.get_per_pixel_mean()