Deep neural network implementation without the learning cliff! This library implements multi-layer perceptrons as a wrapper for the powerful pylearn2 library that's compatible with scikit-learn for a more user-friendly and Pythonic interface. Oh, and it runs on your GPU by default.
NOTE: This project is possible thanks to the nucl.ai Conference on July 20-22. Join us in Vienna!
You'll need to first install some dependencies manually. Unfortunately, pylearn2 isn't yet installable via PyPI and recommends an editable (pip -e) installation:
> git clone https://github.com/lisa-lab/pylearn2.git > cd pylearn2; python setup.py develop
Once that's done, you can grab this repository and set your PYTHONPATH to point to the correct folder. A setup.py file is coming soon for the official version 0.1!
To run a visualization that uses the sknn.mlp.MultiLayerPerceptron just run the following command:
> PYTHONPATH=. python examples/plot_mlp.py --params activation
The datasets are randomized each time, but the output should be an image that looks like this...
- Full tests for sklearn
ClassifierandRegressorcompatibility. - Quick start in the README.rst file showing how to get an estimator.
- Allow using all layer types as hidden layers, not linear only for output.
- Better error checking for the layer specifications, useful messages otherwise.
- Installation using
setup.pylike all normal Python projects!