yangnedu362 / GORU-tensorflow

Gated Orthogonal Recurrent Unit implementation in tensorflow

GORU-tensorflow

Gated Orthogonal Recurrent Unit

If you find this work useful, please cite arXiv:1706.02761.

Installation

requires TensorFlow 1.2.0

Usage

To use GORU in your model, simply copy GORU.py.

Then you can use GORU in the same way you use built-in LSTM:

from GORU import GORUCell
cell = GORUCell(n_hidden, fft=True)

Args:

• n_hidden: Integer.
• capacity: Optional. Integer. Only works for tunable style.
• fft: Optional. Bool. If True, GORU is set to FFT style. Default is True.

Example tasks for GORU

Three tasks for RNN in the paper are shown here. Use -h for more information

Copying Memory Task

python copying_task.py --model GORU

Denoise Task

python denoise_task.py --model GORU

Parenthesis Task

python paren_task.py --model GORU