Metadata-Version: 2.0
Name: phased-lstm-keras
Version: 1.0.2
Summary: Keras implementation of Phased LSTM
Home-page: https://github.com/fferroni/PhasedLSTM-Keras
Author: Francesco Ferroni
Author-email: francescoferroni1@gmail.com
License: GPLv3+
Keywords: machinelearning deeplearning development
Platform: UNKNOWN
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: License :: OSI Approved :: GNU General Public License v3 or later (GPLv3+)
Classifier: Programming Language :: Python :: 2
Classifier: Programming Language :: Python :: 2.7
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.3
Classifier: Programming Language :: Python :: 3.4
Classifier: Programming Language :: Python :: 3.5
Requires-Dist: keras (>=2.0.2)
Requires-Dist: matplotlib (>=2.0.0)
Requires-Dist: numpy (>=1.12.1)

PhasedLSTM-Keras
================

Keras implementation of Phased LSTM [https://arxiv.org/abs/1610.09513], from NIPS 2016.

To install run:
```pip install phased_lstm_keras```

Works both with Theano and Tensorflow backend (tested on Theano 0.9 and Tensorflow 1.0).

Classification performance compared to standard Keras LSTM for MNIST dataset:

.. image:: mnist_plstm_lstm_comparison_acc.png
   :height: 100px
   :width: 100px
   :alt: Accuracy [red: PLSTM, black: LSTM]
   :align: center

.. image:: mnist_plstm_lstm_comparison_loss.png
   :height: 100px
   :width: 100px
   :alt: Loss [red: PLSTM, black: LSTM]
   :align: center


