|
371
|
1 #!/usr/bin/python
|
|
|
2 # coding: utf-8
|
|
|
3
|
|
|
4 import ift6266
|
|
|
5 import pylearn
|
|
|
6
|
|
|
7 import numpy
|
|
|
8 import theano
|
|
|
9 import time
|
|
|
10
|
|
|
11 import pylearn.version
|
|
|
12 import theano.tensor as T
|
|
|
13 from theano.tensor.shared_randomstreams import RandomStreams
|
|
|
14
|
|
|
15 import copy
|
|
|
16 import sys
|
|
|
17 import os
|
|
|
18 import os.path
|
|
|
19
|
|
|
20 from jobman import DD
|
|
|
21 import jobman, jobman.sql
|
|
|
22 from pylearn.io import filetensor
|
|
|
23
|
|
|
24 from utils import produit_cartesien_jobs
|
|
|
25 from copy import copy
|
|
|
26
|
|
|
27 from sgd_optimization import SdaSgdOptimizer
|
|
|
28
|
|
|
29 #from ift6266.utils.scalar_series import *
|
|
|
30 from ift6266.utils.seriestables import *
|
|
|
31 import tables
|
|
|
32
|
|
|
33 from ift6266 import datasets
|
|
|
34 from config import *
|
|
|
35
|
|
|
36 '''
|
|
|
37 Function called by jobman upon launching each job
|
|
|
38 Its path is the one given when inserting jobs: see EXPERIMENT_PATH
|
|
|
39 '''
|
|
|
40 def jobman_entrypoint(state, channel):
|
|
|
41 # record mercurial versions of each package
|
|
|
42 pylearn.version.record_versions(state,[theano,ift6266,pylearn])
|
|
|
43 # TODO: remove this, bad for number of simultaneous requests on DB
|
|
|
44 channel.save()
|
|
|
45
|
|
|
46 # For test runs, we don't want to use the whole dataset so
|
|
|
47 # reduce it to fewer elements if asked to.
|
|
|
48 rtt = None
|
|
|
49 if state.has_key('reduce_train_to'):
|
|
|
50 rtt = state['reduce_train_to']
|
|
|
51 elif REDUCE_TRAIN_TO:
|
|
|
52 rtt = REDUCE_TRAIN_TO
|
|
|
53
|
|
|
54 n_ins = 32*32
|
|
|
55 n_outs = 62 # 10 digits, 26*2 (lower, capitals)
|
|
|
56
|
|
|
57 examples_per_epoch = NIST_ALL_TRAIN_SIZE
|
|
|
58
|
|
|
59 PATH = ''
|
|
|
60 maximum_exemples=int(500000) #Maximum number of exemples seen
|
|
|
61
|
|
|
62
|
|
|
63
|
|
|
64 print "Creating optimizer with state, ", state
|
|
|
65
|
|
|
66 optimizer = SdaSgdOptimizer(dataset=datasets.nist_all(),
|
|
|
67 hyperparameters=state, \
|
|
|
68 n_ins=n_ins, n_outs=n_outs,\
|
|
|
69 examples_per_epoch=examples_per_epoch, \
|
|
|
70 max_minibatches=rtt)
|
|
|
71
|
|
|
72
|
|
|
73
|
|
|
74
|
|
|
75
|
|
|
76 if os.path.exists(PATH+'params_finetune_NIST.txt'):
|
|
|
77 print ('\n finetune = NIST ')
|
|
|
78 optimizer.reload_parameters(PATH+'params_finetune_NIST.txt')
|
|
|
79 print "For" + str(maximum_exemples) + "over the NIST training set: "
|
|
|
80 optimizer.training_error(datasets.nist_all(maxsize=maximum_exemples))
|
|
|
81
|
|
|
82
|
|
|
83 if os.path.exists(PATH+'params_finetune_P07.txt'):
|
|
|
84 print ('\n finetune = P07 ')
|
|
|
85 optimizer.reload_parameters(PATH+'params_finetune_P07.txt')
|
|
|
86 print "For" + str(maximum_exemples) + "over the P07 training set: "
|
|
|
87 optimizer.training_error(datasets.nist_P07(maxsize=maximum_exemples))
|
|
|
88
|
|
|
89
|
|
|
90 if os.path.exists(PATH+'params_finetune_NIST_then_P07.txt'):
|
|
|
91 print ('\n finetune = NIST then P07')
|
|
|
92 optimizer.reload_parameters(PATH+'params_finetune_NIST_then_P07.txt')
|
|
|
93 print "For" + str(maximum_exemples) + "over the NIST training set: "
|
|
|
94 optimizer.training_error(datasets.nist_all(maxsize=maximum_exemples))
|
|
|
95 print "For" + str(maximum_exemples) + "over the P07 training set: "
|
|
|
96 optimizer.training_error(datasets.nist_P07(maxsize=maximum_exemples))
|
|
|
97
|
|
|
98 if os.path.exists(PATH+'params_finetune_P07_then_NIST.txt'):
|
|
|
99 print ('\n finetune = P07 then NIST')
|
|
|
100 optimizer.reload_parameters(PATH+'params_finetune_P07_then_NIST.txt')
|
|
|
101 print "For" + str(maximum_exemples) + "over the P07 training set: "
|
|
|
102 optimizer.training_error(datasets.nist_P07(maxsize=maximum_exemples))
|
|
|
103 print "For" + str(maximum_exemples) + "over the NIST training set: "
|
|
|
104 optimizer.training_error(datasets.nist_all(maxsize=maximum_exemples))
|
|
|
105
|
|
|
106 channel.save()
|
|
|
107
|
|
|
108 return channel.COMPLETE
|
|
|
109
|
|
|
110
|
|
|
111
|
|
|
112 if __name__ == '__main__':
|
|
|
113
|
|
|
114
|
|
|
115 chanmock = DD({'COMPLETE':0,'save':(lambda:None)})
|
|
|
116 jobman_entrypoint(DD(DEFAULT_HP_NIST), chanmock)
|
|
|
117
|
|
|
118
|
