Mercurial > pylearn
changeset 1490:9d2323513092
added formulas/nnet.py with inverse_max_pooling(...)
| author | Emmanuel Bengio <bengioe@iro.umontreal.ca> |
|---|---|
| date | Wed, 27 Jul 2011 10:42:03 -0400 |
| parents | 35a3a4e2d999 |
| children | 8be8cdde97ee |
| files | pylearn/formulas/nnet.py |
| diffstat | 1 files changed, 5 insertions(+), 0 deletions(-) [+] |
line wrap: on
line diff
--- a/pylearn/formulas/nnet.py Wed Jul 27 10:37:41 2011 -0400 +++ b/pylearn/formulas/nnet.py Wed Jul 27 10:42:03 2011 -0400 @@ -22,6 +22,11 @@ The shape of the pooling windows ignore_borders : boolean Will pad borders with zeros if true + + Returns + ------- + ret : 4D tensor + A Theano variable with same shape as output_shape """ # flatten the input and repeat it repeated_input = [max_pool_out.flatten()]*(pooling_shape[0]*pooling_shape[1])