Mercurial > ift6266

comparison AMT/utils.py @ 396:116b2de2c0a4

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
utils for the amazon MT code
author goldfinger
date Tue, 27 Apr 2010 13:47:33 -0400
parents
children
comparison
equal deleted inserted replaced
395:f61a04074723 396:116b2de2c0a4
1 """ This file contains different utility functions that are not connected
2 in anyway to the networks presented in the tutorials, but rather help in
3 processing the outputs into a more understandable way.
4
5 For example ``tile_raster_images`` helps in generating a easy to grasp
6 image from a set of samples or weights.
7 """
8
9
10 import numpy
11
12
13 def scale_to_unit_interval(ndar,eps=1e-8):
14 """ Scales all values in the ndarray ndar to be between 0 and 1 """
15 ndar = ndar.copy()
16 ndar -= ndar.min()
17 ndar *= 1.0 / (ndar.max()+eps)
18 return ndar
19
20
21 def tile_raster_images(X, img_shape, tile_shape,tile_spacing = (0,0),
22 scale_rows_to_unit_interval = True, output_pixel_vals = True):
23 """
24 Transform an array with one flattened image per row, into an array in
25 which images are reshaped and layed out like tiles on a floor.
26
27 This function is useful for visualizing datasets whose rows are images,
28 and also columns of matrices for transforming those rows
29 (such as the first layer of a neural net).
30
31 :type X: a 2-D ndarray or a tuple of 4 channels, elements of which can
32 be 2-D ndarrays or None;
33 :param X: a 2-D array in which every row is a flattened image.
34
35 :type img_shape: tuple; (height, width)
36 :param img_shape: the original shape of each image
37
38 :type tile_shape: tuple; (rows, cols)
39 :param tile_shape: the number of images to tile (rows, cols)
40
41 :param output_pixel_vals: if output should be pixel values (i.e. int8
42 values) or floats
43
44 :param scale_rows_to_unit_interval: if the values need to be scaled before
45 being plotted to [0,1] or not
46
47
48 :returns: array suitable for viewing as an image.
49 (See:`PIL.Image.fromarray`.)
50 :rtype: a 2-d array with same dtype as X.
51
52 """
53
54 assert len(img_shape) == 2
55 assert len(tile_shape) == 2
56 assert len(tile_spacing) == 2
57
58 # The expression below can be re-written in a more C style as
59 # follows :
60 #
61 # out_shape = [0,0]
62 # out_shape[0] = (img_shape[0]+tile_spacing[0])*tile_shape[0] -
63 # tile_spacing[0]
64 # out_shape[1] = (img_shape[1]+tile_spacing[1])*tile_shape[1] -
65 # tile_spacing[1]
66 out_shape = [(ishp + tsp) * tshp - tsp for ishp, tshp, tsp
67 in zip(img_shape, tile_shape, tile_spacing)]
68
69 if isinstance(X, tuple):
70 assert len(X) == 4
71 # Create an output numpy ndarray to store the image
72 if output_pixel_vals:
73 out_array = numpy.zeros((out_shape[0], out_shape[1], 4), dtype='uint8')
74 else:
75 out_array = numpy.zeros((out_shape[0], out_shape[1], 4), dtype=X.dtype)
76
77 #colors default to 0, alpha defaults to 1 (opaque)
78 if output_pixel_vals:
79 channel_defaults = [0,0,0,255]
80 else:
81 channel_defaults = [0.,0.,0.,1.]
82
83 for i in xrange(4):
84 if X[i] is None:
85 # if channel is None, fill it with zeros of the correct
86 # dtype
87 out_array[:,:,i] = numpy.zeros(out_shape,
88 dtype='uint8' if output_pixel_vals else out_array.dtype
89 )+channel_defaults[i]
90 else:
91 # use a recurrent call to compute the channel and store it
92 # in the output
93 out_array[:,:,i] = tile_raster_images(X[i], img_shape, tile_shape, tile_spacing, scale_rows_to_unit_interval, output_pixel_vals)
94 return out_array
95
96 else:
97 # if we are dealing with only one channel
98 H, W = img_shape
99 Hs, Ws = tile_spacing
100
101 # generate a matrix to store the output
102 out_array = numpy.zeros(out_shape, dtype='uint8' if output_pixel_vals else X.dtype)
103
104
105 for tile_row in xrange(tile_shape[0]):
106 for tile_col in xrange(tile_shape[1]):
107 if tile_row * tile_shape[1] + tile_col < X.shape[0]:
108 if scale_rows_to_unit_interval:
109 # if we should scale values to be between 0 and 1
110 # do this by calling the `scale_to_unit_interval`
111 # function
112 this_img = scale_to_unit_interval(X[tile_row * tile_shape[1] + tile_col].reshape(img_shape))
113 else:
114 this_img = X[tile_row * tile_shape[1] + tile_col].reshape(img_shape)
115 # add the slice to the corresponding position in the
116 # output array
117 out_array[
118 tile_row * (H+Hs):tile_row*(H+Hs)+H,
119 tile_col * (W+Ws):tile_col*(W+Ws)+W
120 ] \
121 = this_img * (255 if output_pixel_vals else 1)
122 return out_array
123
124
125