Skip to content

uchihashikenshi/attention_time

Repository files navigation

Attention Neural Network for Time-Series

AttentionalTime is a Python implementation of a time-series model with (optional) attention where the encoder is CNN, decoder is LSTM.

The attention model is from __.

This project is maintained by Kenshi Uchihashi.

Dependencies

Python

  • numpy
  • tensorflow

Quickstart

We are going to working with some example data in data/ folder. It contains delicious dataset, Twitter hashtags, and so on. First run the data-processing code like this:

python preprocessing.py

Details

Preprocessing options(preprocessing.py)

Data options

  • data_dir:

Training options()

Data options

Model options

Optimization options

Other options

Evaluation options()

Visualization options()

Acknowledgements

Our implementation utilizes code from the following:

Licence

About

Python implementation of a time-series model with (optional) attention where the encoder is CNN, decoder is LSTM, and more.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published