GitHub - nglazyrin/chordest-py

1 Branch 0 Tags

Name	Name	Last commit message	Last commit date
Latest commit unknown Append rows when at least 1 row was read (was 2) Mar 11, 2014 40f95de · Mar 11, 2014 History 6 Commits
DL	DL	No corruption when level = 0, binomial corruption by default. Initial…	Mar 11, 2014
Test	Test	Measure test time	Jan 29, 2014
Utils	Utils	Append rows when at least 1 row was read (was 2)	Mar 11, 2014
.gitignore	.gitignore	Changes to make experiments with network layouts easier	Oct 18, 2013
README	README	initial commit	Oct 4, 2013
__init__.py	__init__.py	initial commit	Oct 4, 2013
a.exe	a.exe	initial commit	Oct 4, 2013
base_train_SdA.py	base_train_SdA.py	Changes to make experiments with network layouts easier	Oct 18, 2013
print_weights.py	print_weights.py	Changes to make experiments with network layouts easier	Oct 18, 2013
stats.py	stats.py	Changes to make experiments with network layouts easier	Oct 18, 2013
test_SdA.py	test_SdA.py	Changes to make experiments with network layouts easier	Oct 18, 2013
test_SdA_circ.py	test_SdA_circ.py	Changes to make experiments with network layouts easier	Oct 18, 2013
theano_test.py	theano_test.py	Changes to make experiments with network layouts easier	Oct 18, 2013
train_SdA12.py	train_SdA12.py	Changes to make experiments with network layouts easier	Oct 18, 2013
train_classifier.py	train_classifier.py	Changes to make experiments with network layouts easier	Oct 18, 2013
train_test.py	train_test.py	Changes to make experiments with network layouts easier	Oct 18, 2013
train_test2.py	train_test2.py	Changes to make experiments with network layouts easier	Oct 18, 2013
train_test3.py	train_test3.py	Changes to make experiments with network layouts easier	Oct 18, 2013

Repository files navigation

Here is the Python code that can be used to train a stacked denoising
autoencoder. This code was used to prepare the paper for Notes of Kazan
Federal University in September, 2013, and to get results for the dissertation.