Mercurial > pylearn
view sandbox/rbm/main.py @ 420:040cb796f4e0
Removed feature of passing the file as a pathname from filetensor.{read, write}
Rationale:
1. The common pattern of multiple successive reads to load a list of arrays
doesn't work for the string argument.
2. It isn't clear without checking the docs whether the write() will open a file
in append mode, which could lead to bugs.
3. It's a fairly useless feature anyway, because the truly lazy programmer can
call write(open(filename, 'w'), mat) to reproduce the old behaviour if he
wants it, and without the open-mode ambiguity.
| author | James Bergstra <bergstrj@iro.umontreal.ca> |
|---|---|
| date | Tue, 15 Jul 2008 12:57:21 -0400 |
| parents | c2e6a8fcc35e |
| children |
line wrap: on
line source
#!/usr/bin/python """ Simple SGD RBM training. (An example of how to use the model.) """ import numpy nonzero_instances = [] #nonzero_instances.append({0: 1, 1: 1}) #nonzero_instances.append({0: 1, 2: 1}) nonzero_instances.append({1: 0.1, 5: 0.5, 9: 1}) nonzero_instances.append({2: 0.3, 5: 0.5, 8: 0.8}) nonzero_instances.append({1: 0.2, 2: 0.3, 5: 0.5}) import model model = model.Model(input_dimension=10, hidden_dimension=6) for i in xrange(100000): # Select an instance instance = nonzero_instances[i % len(nonzero_instances)] # SGD update over instance model.update([instance])