Mercurial > ift6266

annotate datasets/defs.py @ 245:0de14b2034c6

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
change in testtransformation to test PNIST pipeline
author Xavier Glorot <glorotxa@iro.umontreal.ca>
date Tue, 16 Mar 2010 12:13:49 -0400
parents f0f47b045cbf
children 476da2ba6a12
rev   line source
181
f0f47b045cbf Remove a stray cast in the FTDataSet code and export the ocr dataset.
Arnaud Bergeron <abergeron@gmail.com>
parents: 180
diff changeset
1 __all__ = ['nist_digits', 'nist_lower', 'nist_upper', 'nist_all', 'ocr']
163
4b28d7382dbf Add inital implementation of datasets.
Arnaud Bergeron <abergeron@gmail.com>
parents:
diff changeset
2
4b28d7382dbf Add inital implementation of datasets.
Arnaud Bergeron <abergeron@gmail.com>
parents:
diff changeset
3 from ftfile import FTDataSet
180
76bc047df5ee Add dtype conversion and rescaling to the read path.
Arnaud Bergeron <abergeron@gmail.com>
parents: 175
diff changeset
4 import theano
163
4b28d7382dbf Add inital implementation of datasets.
Arnaud Bergeron <abergeron@gmail.com>
parents:
diff changeset
5
175
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
6 NIST_PATH = '/data/lisa/data/nist/by_class/'
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
7 DATA_PATH = '/data/lisa/data/ift6266h10/'
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
8
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
9 nist_digits = FTDataSet(train_data = [NIST_PATH+'digits/digits_train_data.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
10 train_lbl = [NIST_PATH+'digits/digits_train_labels.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
11 test_data = [NIST_PATH+'digits/digits_test_data.ft'],
180
76bc047df5ee Add dtype conversion and rescaling to the read path.
Arnaud Bergeron <abergeron@gmail.com>
parents: 175
diff changeset
12 test_lbl = [NIST_PATH+'digits/digits_test_labels.ft'],
76bc047df5ee Add dtype conversion and rescaling to the read path.
Arnaud Bergeron <abergeron@gmail.com>
parents: 175
diff changeset
13 indtype=theano.config.floatX, inscale=255.)
175
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
14 nist_lower = FTDataSet(train_data = [NIST_PATH+'lower/lower_train_data.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
15 train_lbl = [NIST_PATH+'lower/lower_train_labels.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
16 test_data = [NIST_PATH+'lower/lower_test_data.ft'],
180
76bc047df5ee Add dtype conversion and rescaling to the read path.
Arnaud Bergeron <abergeron@gmail.com>
parents: 175
diff changeset
17 test_lbl = [NIST_PATH+'lower/lower_test_labels.ft'],
76bc047df5ee Add dtype conversion and rescaling to the read path.
Arnaud Bergeron <abergeron@gmail.com>
parents: 175
diff changeset
18 indtype=theano.config.floatX, inscale=255.)
175
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
19 nist_upper = FTDataSet(train_data = [NIST_PATH+'upper/upper_train_data.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
20 train_lbl = [NIST_PATH+'upper/upper_train_labels.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
21 test_data = [NIST_PATH+'upper/upper_test_data.ft'],
180
76bc047df5ee Add dtype conversion and rescaling to the read path.
Arnaud Bergeron <abergeron@gmail.com>
parents: 175
diff changeset
22 test_lbl = [NIST_PATH+'upper/upper_test_labels.ft'],
76bc047df5ee Add dtype conversion and rescaling to the read path.
Arnaud Bergeron <abergeron@gmail.com>
parents: 175
diff changeset
23 indtype=theano.config.floatX, inscale=255.)
163
4b28d7382dbf Add inital implementation of datasets.
Arnaud Bergeron <abergeron@gmail.com>
parents:
diff changeset
24
175
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
25 nist_all = FTDataSet(train_data = [DATA_PATH+'train_data.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
26 train_lbl = [DATA_PATH+'train_labels.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
27 test_data = [DATA_PATH+'test_data.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
28 test_lbl = [DATA_PATH+'test_labels.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
29 valid_data = [DATA_PATH+'valid_data.ft'],
180
76bc047df5ee Add dtype conversion and rescaling to the read path.
Arnaud Bergeron <abergeron@gmail.com>
parents: 175
diff changeset
30 valid_lbl = [DATA_PATH+'valid_labels.ft'],
76bc047df5ee Add dtype conversion and rescaling to the read path.
Arnaud Bergeron <abergeron@gmail.com>
parents: 175
diff changeset
31 indtype=theano.config.floatX, inscale=255.)
175
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
32
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
33 ocr = FTDataSet(train_data = [DATA_PATH+'ocr_train_data.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
34 train_lbl = [DATA_PATH+'ocr_train_labels.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
35 test_data = [DATA_PATH+'ocr_test_data.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
36 test_lbl = [DATA_PATH+'ocr_test_labels.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
37 valid_data = [DATA_PATH+'ocr_valid_data.ft'],
224321bf043a Define the ocr dataset and use the existing split for nist.
Arnaud Bergeron <abergeron@gmail.com>
parents: 164
diff changeset
38 valid_lbl = [DATA_PATH+'ocr_valid_labels.ft'])