Mercurial > ift6266

annotate deep/crbm/utils.py @ 339:ffbf0e41bcee

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
author fsavard
date Sat, 17 Apr 2010 20:29:18 -0400
parents
children 82dae7c46046
rev   line source
339
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
1 #!/usr/bin/python
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
2 # coding: utf-8
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
3
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
4 from __future__ import with_statement
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
5
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
6 from jobman import DD
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
7
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
8 from pylearn.io.seriestables import *
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
9 import tables
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
10
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
11
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
12
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
13 # from pylearn codebase
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
14 # useful in __init__(param1, param2, etc.) to save
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
15 # values in self.param1, self.param2... just call
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
16 # update_locals(self, locals())
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
17 def update_locals(obj, dct):
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
18 if 'self' in dct:
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
19 del dct['self']
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
20 obj.__dict__.update(dct)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
21
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
22 # from a dictionary of possible values for hyperparameters, e.g.
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
23 # hp_values = {'learning_rate':[0.1, 0.01], 'num_layers': [1,2]}
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
24 # create a list of other dictionaries representing all the possible
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
25 # combinations, thus in this example creating:
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
26 # [{'learning_rate': 0.1, 'num_layers': 1}, ...]
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
27 # (similarly for combinations (0.1, 2), (0.01, 1), (0.01, 2))
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
28 def produit_cartesien_jobs(val_dict):
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
29 job_list = [DD()]
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
30 all_keys = val_dict.keys()
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
31
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
32 for key in all_keys:
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
33 possible_values = val_dict[key]
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
34 new_job_list = []
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
35 for val in possible_values:
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
36 for job in job_list:
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
37 to_insert = job.copy()
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
38 to_insert.update({key: val})
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
39 new_job_list.append(to_insert)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
40 job_list = new_job_list
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
41
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
42 return job_list
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
43
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
44 def jobs_from_reinsert_list(cols, job_vals):
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
45 job_list = []
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
46 for vals in job_vals:
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
47 job = DD()
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
48 for i, col in enumerate(cols):
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
49 job[col] = vals[i]
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
50 job_list.append(job)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
51
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
52 return job_list
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
53
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
54 def save_params(all_params, filename):
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
55 import pickle
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
56 with open(filename, 'wb') as f:
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
57 values = [p.value for p in all_params]
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
58
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
59 # -1 for HIGHEST_PROTOCOL
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
60 pickle.dump(values, f, -1)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
61
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
62 # Perform insertion into the Postgre DB based on combination
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
63 # of hyperparameter values above
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
64 # (see comment for produit_cartesien_jobs() to know how it works)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
65 def jobman_insert_job_vals(job_db, experiment_path, job_vals):
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
66 jobs = produit_cartesien_jobs(job_vals)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
67
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
68 db = jobman.sql.db(job_db)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
69 for job in jobs:
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
70 job.update({jobman.sql.EXPERIMENT: experiment_path})
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
71 jobman.sql.insert_dict(job, db)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
72
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
73 def jobman_insert_specific_jobs(job_db, experiment_path,
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
74 insert_cols, insert_vals):
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
75 jobs = jobs_from_reinsert_list(insert_cols, insert_vals)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
76
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
77 db = jobman.sql.db(job_db)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
78 for job in jobs:
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
79 job.update({jobman.sql.EXPERIMENT: experiment_path})
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
80 jobman.sql.insert_dict(job, db)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
81
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
82 # Just a shortcut for a common case where we need a few
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
83 # related Error (float) series
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
84 def get_accumulator_series_array( \
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
85 hdf5_file, group_name, series_names,
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
86 reduce_every,
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
87 index_names=('epoch','minibatch'),
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
88 stdout_too=True,
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
89 skip_hdf5_append=False):
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
90 all_series = []
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
91
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
92 new_group = hdf5_file.createGroup('/', group_name)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
93
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
94 other_targets = []
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
95 if stdout_too:
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
96 other_targets = [StdoutAppendTarget()]
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
97
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
98 for sn in series_names:
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
99 series_base = \
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
100 ErrorSeries(error_name=sn,
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
101 table_name=sn,
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
102 hdf5_file=hdf5_file,
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
103 hdf5_group=new_group._v_pathname,
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
104 index_names=index_names,
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
105 other_targets=other_targets,
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
106 skip_hdf5_append=skip_hdf5_append)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
107
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
108 all_series.append( \
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
109 AccumulatorSeriesWrapper( \
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
110 base_series=series_base,
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
111 reduce_every=reduce_every))
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
112
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
113 ret_wrapper = SeriesArrayWrapper(all_series)
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
114
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
115 return ret_wrapper
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
116
ffbf0e41bcee Aded code to run experiment on cluster, separate configuration from other machinery. Not tested yet.
fsavard
parents:
diff changeset
117