search

lmjohns3 / theanets

Neural network toolkit for Python
http://theanets.rtfd.org
MIT License
328 stars 73 forks source link

No Diagnostic message with SGD #19

Closed kadarakos closed 10 years ago

kadarakos commented 10 years ago

I use sgd for training and I don't get the diagnostic messages of the trainer for some reason and I do not know what am I doing wrong: e = Experiment(Regressor, layers=(input_layer, hidden1, output_layer), train_batches = 100, optimize='sgd')

lmjohns3 commented 10 years ago

Ah, you probably need to enable logging for your program. (By default the Python logging package does not show any output on stdout or stderr.) You can do this in any number of ways, but I think the easiest is:

import climate
climate.enable_default_logging()

The climate package is a small wrapper around the logging package; it should already be installed along with theanets. The above lines are in examples/utils.py, which is why the example scripts show the logging output.

kadarakos commented 10 years ago

Oh nice, thank you!

2014-04-11 9:51 GMT-04:00 Leif Johnson notifications@github.com:

Closed #19 https://github.com/lmjohns3/theano-nets/issues/19.

Reply to this email directly or view it on GitHubhttps://github.com/lmjohns3/theano-nets/issues/19 .