-
Notifications
You must be signed in to change notification settings - Fork 1
/
jeffrey.py
310 lines (257 loc) · 11.3 KB
/
jeffrey.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
#!/usr/bin/env python2.7
# encoding: utf-8
'''
'''
from __future__ import print_function
# Load jeffreys model and try to make some predictions
import sys
sys.path.append('../kaggle_diabetic_retinopathy/')
import cPickle as pickle
import re
import glob
import os
import time
import lasagne as nn
from lasagne.layers import dnn
from lasagne.nonlinearities import LeakyRectify
import theano
import theano.tensor as T
import numpy as np
import pandas as pd
from generators import DataLoader
from layers import ApplyNonlinearity
from utils import hms, architecture_string, get_img_ids_from_iter
def from_raw():
layers = []
batch_size = 2
num_channels = 3
input_height = input_width = 512
leakiness = 0.5
l_in_imgdim = nn.layers.InputLayer(
shape=(batch_size, 2),
name='imgdim'
)
l_in1 = nn.layers.InputLayer(
shape=(batch_size, num_channels, input_width, input_height),
name='images'
)
layers.append(l_in1)
Conv2DLayer = dnn.Conv2DDNNLayer
MaxPool2DLayer = dnn.MaxPool2DDNNLayer
DenseLayer = nn.layers.DenseLayer
l_conv = Conv2DLayer(layers[-1],
num_filters=32, filter_size=(7, 7), stride=(2, 2),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_pool = MaxPool2DLayer(layers[-1], pool_size=(3, 3), stride=(2, 2))
layers.append(l_pool)
l_conv = Conv2DLayer(layers[-1],
num_filters=32, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_conv = Conv2DLayer(layers[-1],
num_filters=32, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_pool = MaxPool2DLayer(layers[-1], pool_size=(3, 3), stride=(2, 2))
layers.append(l_pool)
l_conv = Conv2DLayer(layers[-1],
num_filters=64, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_conv = Conv2DLayer(layers[-1],
num_filters=64, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_pool = MaxPool2DLayer(layers[-1], pool_size=(3, 3), stride=(2, 2))
layers.append(l_pool)
l_conv = Conv2DLayer(layers[-1],
num_filters=128, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_conv = Conv2DLayer(layers[-1],
num_filters=128, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_conv = Conv2DLayer(layers[-1],
num_filters=128, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_conv = Conv2DLayer(layers[-1],
num_filters=128, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_pool = MaxPool2DLayer(layers[-1], pool_size=(3, 3), stride=(2, 2))
layers.append(l_pool)
l_conv = Conv2DLayer(layers[-1],
num_filters=256, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_conv = Conv2DLayer(layers[-1],
num_filters=256, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_conv = Conv2DLayer(layers[-1],
num_filters=256, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_conv = Conv2DLayer(layers[-1],
num_filters=256, filter_size=(3, 3), stride=(1, 1),
pad='same',
nonlinearity=LeakyRectify(leakiness),
W=nn.init.Orthogonal(1.0), b=nn.init.Constant(0.1),
untie_biases=True)
layers.append(l_conv)
l_pool = MaxPool2DLayer(layers[-1], pool_size=(3, 3), stride=(2, 2),
name='coarse_last_pool')
layers.append(l_pool)
layers.append(nn.layers.DropoutLayer(layers[-1], p=0.5))
layers.append(DenseLayer(layers[-1],
nonlinearity=None,
num_units=1024,
W=nn.init.Orthogonal(1.0),
b=nn.init.Constant(0.1),
name='first_fc_0'))
l_pool = nn.layers.FeaturePoolLayer(layers[-1],
pool_size=2,
pool_function=T.max)
layers.append(l_pool)
l_first_repr = layers[-1]
l_coarse_repr = nn.layers.concat([l_first_repr,
l_in_imgdim])
layers.append(l_coarse_repr)
# Combine representations of both eyes.
layers.append(
nn.layers.ReshapeLayer(layers[-1], shape=(batch_size // 2, -1)))
layers.append(nn.layers.DropoutLayer(layers[-1], p=0.5))
layers.append(nn.layers.DenseLayer(layers[-1],
nonlinearity=None,
num_units=1024,
W=nn.init.Orthogonal(1.0),
b=nn.init.Constant(0.1),
name='combine_repr_fc'))
l_pool = nn.layers.FeaturePoolLayer(layers[-1],
pool_size=2,
pool_function=T.max)
layers.append(l_pool)
l_hidden = nn.layers.DenseLayer(nn.layers.DropoutLayer(layers[-1], p=0.5),
num_units=output_dim * 2,
nonlinearity=None, # No softmax yet!
W=nn.init.Orthogonal(1.0),
b=nn.init.Constant(0.1))
layers.append(l_hidden)
# Reshape back to 5.
layers.append(nn.layers.ReshapeLayer(layers[-1],
shape=(batch_size, 5)))
# Apply softmax.
l_out = ApplyNonlinearity(layers[-1],
nonlinearity=nn.nonlinearities.softmax)
layers.append(l_out)
l_ins = [l_in1, l_in_imgdim]
nn.layers.set_all_param_values(l_out, pickle.load(open(RAW_DUMP_PATH, 'rb')))
return l_out, l_ins
DUMP_PATH = '../kaggle_diabetic_retinopathy/dumps/2015_07_17_123003.pkl'
RAW_DUMP_PATH = '../kaggle_diabetic_retinopathy/dumps/2015_07_17_123003_PARAMSDUMP.pkl'
NEW_DUMP_PATH = '../kaggle_diabetic_retinopathy/dumps/2015_07_17_123003.updated.pkl'
IMG_DIR = '../train_ds2_crop/'
model_data = pickle.load(open(DUMP_PATH, 'r'))
train_labels = pd.read_csv(os.path.join('../kaggle_diabetic_retinopathy/data/trainLabels.csv'))
batch_size = model_data['batch_size']
output_dim = 5
chunk_size = model_data['chunk_size']
patient_ids = sorted(set(get_img_ids_from_iter(train_labels.image)))
no_transfo_params = model_data['data_loader_params']['no_transfo_params']
if 'paired_transfos' in model_data:
paired_transfos = model_data['paired_transfos']
else:
paired_transfos = False
# Overwrite the model w/ the updated model (one that is suitable for use w/ newer versions of lasagne).
l_out, l_ins = from_raw()
model_data['l_out'] = l_out
model_data['l_ins'] = l_ins
with open(NEW_DUMP_PATH, 'wb') as fh:
pickle.dump(model_data, fh, protocol=-1)
print('Dumped new model')
output = nn.layers.get_output(l_out, deterministic=True)
input_ndims = [len(nn.layers.get_output_shape(l_in))
for l_in in l_ins]
xs_shared = [nn.utils.shared_empty(dim=ndim)
for ndim in input_ndims]
idx = T.lscalar('idx')
givens = {}
for l_in, x_shared in zip(l_ins, xs_shared):
givens[l_in.input_var] = x_shared[idx * batch_size:(idx + 1) * batch_size]
compute_output = theano.function(
[idx],
output,
givens=givens,
on_unused_input='ignore'
)
data_loader = DataLoader()
new_dataloader_params = model_data['data_loader_params']
new_dataloader_params.update({'images_test': patient_ids})
new_dataloader_params.update({'labels_test': train_labels.level.values})
new_dataloader_params.update({'prefix_train': IMG_DIR})
data_loader.set_params(new_dataloader_params)
num_chunks = int(np.ceil((2 * len(patient_ids)) / float(chunk_size)))
def do_pred(img_ids):
test_gen = lambda: data_loader.create_fixed_gen(
img_ids,
chunk_size=chunk_size,
prefix_train=IMG_DIR,
prefix_test=IMG_DIR,
transfo_params=no_transfo_params,
paired_transfos=paired_transfos,
)
outputs = []
for e, (xs_chunk, chunk_shape, chunk_length) in enumerate(test_gen()):
num_batches_chunk = int(np.ceil(chunk_length / float(batch_size)))
print("Chunk %i/%i" % (e + 1, num_chunks))
print(" load data onto GPU")
for x_shared, x_chunk in zip(xs_shared, xs_chunk):
x_shared.set_value(x_chunk)
print(" compute output in batches")
outputs_chunk = []
for b in xrange(num_batches_chunk):
out = compute_output(b)
outputs_chunk.append(out)
outputs_chunk = np.vstack(outputs_chunk)
outputs_chunk = outputs_chunk[:chunk_length]
outputs.append(outputs_chunk)
return np.vstack(outputs), xs_chunk