Especify unsupervised and supervised learning on same yaml file

[Luiz Gustavo Hafemann]

Hey there,

I have been exploring pylearn2 for a couple of weeks, and it has been very helpful for me to get started in my research. I have a question though, on what's the best way to perform unsupervised and supervised training:

I am using Gaussian-RBM for unsupervised pre-training on a dataset, and using its weights to initialize a simple NN (with a "PretrainedLayer" using the learned RBM and a "Softmax" layer on top of it)

So far I have been using two yaml files: one for the RBM, and one for the NN finetuning, and calling the train.py script twice for doing the training.

Is there a way to define unsupervised and supervised learning on the same yaml file? 

I see in "train.py" that you can iterate through "Training phases", but I am not sure how that is modeled in the yaml file.

Thanks in advance,

Gustavo

[David Warde-Farley]

If you make the whole thing be a list of two Train objects in [], train.py will execute them in sequence.

--
 
---
You received this message because you are subscribed to the Google Groups "pylearn-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pylearn-dev...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Luiz Gustavo Hafemann]

I tested it here, and it worked properly.

Thanks David.

Gustavo

Atenciosamente,

Luiz Gustavo Hafemann

--
 
---
You received this message because you are subscribed to a topic in the Google Groups "pylearn-dev" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/pylearn-dev/DGud0hBFfd8/unsubscribe?hl=en-US.
To unsubscribe from this group and all its topics, send an email to pylearn-dev...@googlegroups.com.