Scikit-compatible deep learning framework?

I'm looking for a deep learning framework that is compatible with scikit-learn. Is there anything like this please?

I mainly need:

1- deep believe networks.

2- Stacked Denoising Autoencoders

asked Apr 27 '14 at 06:10

Alex Twain
61●8●9●12

edited Apr 27 '14 at 06:14

Scikit-learn itself got RBMs http://scikit-learn.org/stable/auto_examples/plot_rbm_logistic_classification.html

(Apr 27 '14 at 08:30) eder

@eder yeah but not believe networks and stacked dA.

(Apr 27 '14 at 10:34) Alex Twain

One Answer:

oldest newest most voted

nolearn might be interesting for you. It has an interface mostly compatible to scikit-learn and wraps Decaf and George Dahl’s DBN implementation gdbn. This means that it comes with pretrained convolutional nets and deep belief networks.

Disclaimer: I haven't used it extensively

answered Apr 28 '14 at 10:05

frisbee
1

Subscription:

Once you sign in you will be able to subscribe for any updates here

Tags:

deep-learning ×128
deep-belief-networks ×33
scikit-learn ×21

Asked: Apr 27 '14 at 06:10

Seen: 1,556 times

Last updated: Apr 28 '14 at 10:05

Scikit-compatible deep learning framework?

Subscription:

Related questions