Tensorflow implementation of char-rnn
Python
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
app fixed some typos and updated default values Oct 10, 2016
.gitignore Added option to specify device and added input folder to .gitignore Aug 29, 2016
README.md

README.md

tf-char-rnn

TensorFlow implementation of char-rnn using Multi-layer Recurrent Neural Networks.

Usage

  • Install TensorFlow
  • Put the data (to train on) in a file called 'data.txt'
  • Make model folder by running mkdir model
  • Run python train.py --input_dir=path_to_folder_containing_data.txt
  • For example, if data is in input folder, run python train.py --input_dir=input
  • To generate text using a checkpointed model, run python generate.py
  • All supported arguments can be seen in here

Standing on the shoulder of giants