search

lisa-lab / pylearn2

Warning: This project does not have any current developer. See bellow.
BSD 3-Clause "New" or "Revised" License
2.75k stars 1.09k forks source link

Implement sparse autoencoders with reconstruction sampling #47

Closed goodfeli closed 11 years ago

goodfeli commented 12 years ago

Sprint assignees:

lamblin commented 11 years ago

@yaoli, @thiboeri, can you confirm it has been done?

yaoli commented 11 years ago

Yes.

On Jun 18, 2013, at 6:07 PM, lamblin notifications@github.com wrote:

@yaoli, @thiboeri, can you confirm it has been done?

— Reply to this email directly or view it on GitHub.

lamblin commented 11 years ago

Thanks! Closing.

dwf commented 11 years ago

Is this actually integrated into pylearn2? This is news to me. I don't have a laptop handy but I would like to know where it is. If it isn't in the repo then this ticket shouldn't be closed. On 2013-06-18 7:12 PM, "lamblin" notifications@github.com wrote:

Thanks! Closing.

— Reply to this email directly or view it on GitHubhttps://github.com/lisa-lab/pylearn2/issues/47#issuecomment-19649890 .

lamblin commented 11 years ago

The model, SparseDenoisingAutoencoder is in pylearn2/models/sparse_autoencoder.py. The corresponding costs, SampledMeanBinaryCrossEntropy and SampledMeanSquaredReconstructionError are in pylearn2/costs/autoencoder.py.

goodfeli commented 11 years ago

It's in the repo

https://github.com/lisa-lab/pylearn2/blob/master/pylearn2/models/sparse_autoencoder.py

On Wednesday, June 19, 2013, David Warde-Farley wrote:

Is this actually integrated into pylearn2? This is news to me. I don't have a laptop handy but I would like to know where it is. If it isn't in the repo then this ticket shouldn't be closed. On 2013-06-18 7:12 PM, "lamblin" <notifications@github.com<javascript:_e({}, 'cvml', 'notifications@github.com');>> wrote:

Thanks! Closing.

— Reply to this email directly or view it on GitHub< https://github.com/lisa-lab/pylearn2/issues/47#issuecomment-19649890> .

— Reply to this email directly or view it on GitHubhttps://github.com/lisa-lab/pylearn2/issues/47#issuecomment-19689090 .

Sent from Gmail Mobile

dwf commented 11 years ago

Okay, hadn't seen that. Thanks. On 2013-06-19 11:09 AM, "lamblin" notifications@github.com wrote:

The model, SparseDenoisingAutoencoder is in pylearn2/models/sparse_autoencoder.py. The corresponding costs, SampledMeanBinaryCrossEntropy and SampledMeanSquaredReconstructionError are in pylearn2/costs/autoencoder.py .

— Reply to this email directly or view it on GitHubhttps://github.com/lisa-lab/pylearn2/issues/47#issuecomment-19690154 .