ekzhang / char-rnn-keras

TensorFlow implementation of multi-layer recurrent neural networks for training and sampling from texts

char-rnn-keras

Multi-layer recurrent neural networks for training and sampling from texts, inspired by Andrej Karpathy's article and the original torch source code karpathy/char-rnn.

Requirements

This code is written in Python 3, and it requires the Keras deep learning library.

Input data

All input data should be placed in the data/ directory. Sample training texts are provided.

Usage

To train the model with default settings:

$ python train.py --input tiny-shakespeare.txt

To sample the model at epoch 100:

$ python sample.py 100

Training loss/accuracy is stored in logs/training_log.csv. Model results, including intermediate model weights during training, are stored in the model directory. These are also used by sample.py for sampling.