-
Notifications
You must be signed in to change notification settings - Fork 3
/
util.py
367 lines (340 loc) · 12.5 KB
/
util.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
import numpy as np
import matplotlib.pylab as plt
import torch
from copy import deepcopy
def plot_matrices(
matrix_list,
shape = None,
images_per_row = 10,
scale_limit = None,
figsize = (20, 8),
x_axis_list = None,
filename = None,
title = None,
subtitles = [],
highlight_bad_values = True,
plt = None,
pdf = None,
):
"""Plot the images for each matrix in the matrix_list."""
import matplotlib
from matplotlib import pyplot as plt
fig = plt.figure(figsize = figsize)
fig.set_canvas(plt.gcf().canvas)
if title is not None:
fig.suptitle(title, fontsize = 18, horizontalalignment = 'left', x=0.1)
num_matrixs = len(matrix_list)
rows = np.ceil(num_matrixs / float(images_per_row))
try:
matrix_list_reshaped = np.reshape(np.array(matrix_list), (-1, shape[0],shape[1])) \
if shape is not None else np.array(matrix_list)
except:
matrix_list_reshaped = matrix_list
if scale_limit == "auto":
scale_min = np.Inf
scale_max = -np.Inf
for matrix in matrix_list:
scale_min = min(scale_min, np.min(matrix))
scale_max = max(scale_max, np.max(matrix))
scale_limit = (scale_min, scale_max)
for i in range(len(matrix_list)):
ax = fig.add_subplot(rows, images_per_row, i + 1)
image = matrix_list_reshaped[i].astype(float)
if len(image.shape) == 1:
image = np.expand_dims(image, 1)
if highlight_bad_values:
cmap = matplotlib.cm.binary
cmap.set_bad('red', alpha = 0.2)
mask_key = []
mask_key.append(np.isnan(image))
mask_key.append(np.isinf(image))
mask_key = np.any(np.array(mask_key), axis = 0)
image = np.ma.array(image, mask = mask_key)
else:
cmap = matplotlib.cm.binary
if scale_limit is None:
ax.matshow(image, cmap = cmap)
else:
assert len(scale_limit) == 2, "scale_limit should be a 2-tuple!"
ax.matshow(image, cmap = cmap, vmin = scale_limit[0], vmax = scale_limit[1])
if len(subtitles) > 0:
ax.set_title(subtitles[i])
try:
xlabel = "({0:.4f},{1:.4f})\nshape: ({2}, {3})".format(np.min(image), np.max(image), image.shape[0], image.shape[1])
if x_axis_list is not None:
xlabel += "\n{0}".format(x_axis_list[i])
plt.xlabel(xlabel)
except:
pass
plt.xticks(np.array([]))
plt.yticks(np.array([]))
if filename is not None:
plt.tight_layout()
plt.savefig(filename)
if pdf is not None:
pdf.savefig() # saves the current figure into a pdf page
plt.close()
else:
plt.show()
if scale_limit is not None:
print("scale_limit: ({0:.6f}, {1:.6f})".format(scale_limit[0], scale_limit[1]))
print()
class Early_Stopping(object):
def __init__(self, patience = 100, epsilon = 0, mode = "min"):
self.patience = patience
self.epsilon = epsilon
self.mode = "min"
self.best_value = None
self.wait = 0
def monitor(self, value):
to_stop = False
if self.patience is not None:
if self.best_value is None:
self.best_value = value
self.wait = 0
else:
if (self.mode == "min" and value < self.best_value - self.epsilon) or \
(self.mode == "max" and value > self.best_value + self.epsilon):
self.best_value = value
self.wait = 0
else:
if self.wait >= self.patience:
to_stop = True
else:
self.wait += 1
return to_stop
def record_data(data_record_dict, data_list, key_list):
"""Record data to the dictionary data_record_dict. It records each key: value pair in the corresponding location of
key_list and data_list into the dictionary."""
assert len(data_list) == len(key_list), "the data_list and key_list should have the same length!"
for data, key in zip(data_list, key_list):
if key not in data_record_dict:
data_record_dict[key] = [data]
else:
data_record_dict[key].append(data)
def make_dir(filename):
import os
import errno
if not os.path.exists(os.path.dirname(filename)):
print("directory {0} does not exist, created.".format(os.path.dirname(filename)))
try:
os.makedirs(os.path.dirname(filename))
except OSError as exc: # Guard against race condition
if exc.errno != errno.EEXIST:
print(exc)
raise
def norm(matrix, noise_mode, mode = "L1"):
# Clamp the matrix if noise_mode is "permutation":
if noise_mode == "permutation":
matrix = matrix.clamp(1e-6, 1)
if mode == "L1":
return matrix.abs().mean()
elif mode == "L2":
return torch.sqrt((matrix ** 2).mean() + 1e-9)
elif mode == "info":
return torch.log2(1 + 1 / matrix ** 2).sum() / 2
elif isinstance(mode, tuple):
if mode[0] == "order":
order = mode[1]
return (matrix.abs() ** order).mean()
elif mode[0] == "exp_order":
order = mode[1]
return (torch.exp(matrix.abs() ** order)).mean()
elif mode[0] == "log_order":
order = mode[1]
return (torch.log(matrix.abs() ** order + 1e-9)).mean()
else:
raise
else:
raise Exception("mode not recognized!".format(mode))
def train_test_split(X, y, test_size = 0.1):
import torch
if len(X.shape) == 4:
X = X.view(-1, *X.shape[2:])
y = y.view(-1, *y.shape[2:])
num_examples = len(X)
if test_size is not None:
num_test = int(num_examples * test_size)
num_train = num_examples - num_test
idx_train = np.random.choice(range(num_examples), size = num_train, replace = False)
idx_test = set(range(num_examples)) - set(idx_train)
device = torch.device("cuda" if X.is_cuda else "cpu")
idx_train = torch.LongTensor(list(idx_train)).to(device)
idx_test = torch.LongTensor(list(idx_test)).to(device)
X_train = X[idx_train]
y_train = y[idx_train]
X_test = X[idx_test]
y_test = y[idx_test]
else:
X_train, X_test = X, X
y_train, y_test = y, y
return (X_train, y_train), (X_test, y_test)
def new_dict(Dict, new_content_dict):
from copy import deepcopy
new_Dict = deepcopy(Dict)
new_Dict.update(new_content_dict)
return new_Dict
def format_list(List, interval = "\t", decimals = None):
if decimals is None:
return interval.join(["{0}".format(element) for element in List])
else:
return interval.join(["{0:.{1}f}".format(element, decimals) for element in List])
def sort_two_lists(list1, list2, reverse = False):
from operator import itemgetter
if reverse:
List = deepcopy([list(x) for x in zip(*sorted(zip(deepcopy(list1), deepcopy(list2)), key=itemgetter(0), reverse=True))])
else:
List = deepcopy([list(x) for x in zip(*sorted(zip(deepcopy(list1), deepcopy(list2)), key=itemgetter(0)))])
if len(List) == 0:
return [], []
else:
return List[0], List[1]
def get_args(arg, arg_id = 1, type = "str"):
try:
get_ipython().run_line_magic('matplotlib', 'inline')
arg_return = arg
except:
import sys
try:
arg_return = sys.argv[arg_id]
if type == "int":
arg_return = int(arg_return)
elif type == "float":
arg_return = float(arg_return)
elif type == "bool":
arg_return = eval(arg_return)
elif type == "eval":
arg_return = eval(arg_return)
elif type == "tuple":
if arg_return[0] not in ["(", "["]:
arg_return = eval(arg_return)
else:
splitted = arg_return[1:-1].split(",")
List = []
for item in splitted:
try:
item = eval(item)
except:
pass
List.append(item)
arg_return = tuple(List)
elif type == "str":
pass
else:
raise Exception("type {0} not recognized!".format(type))
except:
# raise
arg_return = arg
return arg_return
def normalize_tensor(X, y, normalize):
import torch
if isinstance(X, np.ndarray):
assert len(X.shape) >= 3
assert len(y.shape) >= 3
XY = np.concatenate([X, y], -2)
if normalize == 0:
pass
elif normalize == 1:
mean = XY.mean()
std = XY.std()
X = (X - mean) / std
y = (y - mean) / std
elif normalize == 2:
shape = tuple(XY.shape)
N = shape[-1]
XY_reshape = XY.reshape(-1, N)
if len(shape) == 3:
mean = XY_reshape.mean(0).reshape(1, 1, N)
std = XY_reshape.std(0).reshape(1, 1, N)
elif len(shape) == 4:
mean = XY_reshape.mean(0).reshape(1, 1, 1, N)
std = XY_reshape.std(0).reshape(1, 1, 1, N)
else:
raise
X = (X - mean) / std
y = (y - mean) / std
elif normalize == 3:
N = XY.shape[-1]
X_new = []
Y_new = []
for i in range(N):
xy_ele = XY[...,i]
x_ele = X[...,i]
y_ele = y[...,i]
xy_max = xy_ele.max()
xy_min = xy_ele.min()
x_new = (x_ele - xy_min) / (xy_max -xy_min)
y_new = (y_ele - xy_min) / (xy_max -xy_min)
X_new.append(x_new)
Y_new.append(y_new)
X = np.stack(X_new, -1)
y = np.stack(Y_new, -1)
elif normalize == 4:
# Make each (...,K,N) zero mean:
X_shape = X.shape
X = X.reshape(-1, *X_shape[-2:])
X = X - X.mean(0, keepdims = True)
X = X.reshape(*X_shape)
# Make each (...,K,N) zero mean and unit std:
y_shape = y.shape
y = y.reshape(-1, *y_shape[-2:])
y = (y - y.mean(0, keepdims = True)) / y.std(0, keepdims = True)
y = y.reshape(*y_shape)
else:
raise
else:
assert len(X.size()) >= 3
assert len(y.size()) >= 3
XY = torch.cat([X, y], -2)
if normalize == 0:
pass
elif normalize == 1:
mean = XY.mean()
std = XY.std()
X = (X - mean) / std
y = (y - mean) / std
elif normalize == 2:
shape = tuple(XY.shape)
N = shape[-1]
XY_reshape = XY.reshape(-1, N)
if len(shape) == 3:
mean = XY_reshape.mean(0).view(1, 1, N)
std = XY_reshape.std(0).view(1, 1, N)
elif len(shape) == 4:
mean = XY_reshape.mean(0).view(1, 1, 1, N)
std = XY_reshape.std(0).view(1, 1, 1, N)
else:
raise
X = (X - mean) / (std + 1e-9)
y = (y - mean) / (std + 1e-9)
elif normalize == 3:
N = XY.shape[-1]
X_new = []
Y_new = []
for i in range(N):
xy_ele = XY[...,i]
x_ele = X[...,i]
y_ele = y[...,i]
xy_max = xy_ele.max()
xy_min = xy_ele.min()
x_new = (x_ele - xy_min) / (xy_max -xy_min)
y_new = (y_ele - xy_min) / (xy_max -xy_min)
X_new.append(x_new)
Y_new.append(y_new)
X = torch.stack(X_new, -1)
y = torch.stack(Y_new, -1)
elif normalize == 4:
pass
# # Make each (...,K,N) zero mean:
# X_shape = X.shape
# X = X.view(-1, *X_shape[-2:])
# X = X - X.mean(0, keepdim = True)
# X = X.view(*X_shape)
# # Make each (...,K,N) zero mean and unit std:
# y_shape = y.shape
# y = y.view(-1, *y_shape[-2:])
# y = (y - y.mean(0, keepdim = True)) / y.std(0, keepdim = True)
# y = y.view(*y_shape)
else:
raise
return X, y