/ntsa

Neural Time Series Analysis

Primary LanguageJupyter NotebookMIT LicenseMIT

Neural Time Series Analysis

Introduction

Neural Time Series Anaylsis is an attempt to provide an easy to use interface to modern time-series research from the machine learning community. The library features several models, for 1d time-series regression that have been collected thanks to researchers that published their code and hackers that tried to reproduce them.

Features

  1. Recurrent models can use the Input Attention mechanism as developed in the DARNN paper, to increate interpretability in case of high-dimensional feature space.
  2. (Conditional) Neural Processes in the regression setting, can provide mean and variance for each estimate.
  3. Soft Dynamic Time Warping for time-series alignment

Datasets

This library comes with the following datasets:

Pickle file can be found here.

To adapt a new dataset, please check utils.dataset_utils.py.

Models

Install

For Pip users:

Pip

git clone github.com/d3sm0/ntsa.git
cd ntsa
pip install -r requirements.txt

This library also requires to install the soft-dtw loss function.

Docker

If you are used to docker than maybe this can work better:

make docker
make dev

Usage

python main.py --model seq2seq --loss sdtw  --mode train --data_path data/

Remark

This repo is still under refinement and I'm planning to move to eager mode in the upcoming weeks. Nevertheless is usable for early experiments, there for Fork it and hack with it :) Pull request are welcome, and if you need help in making it work, just open an issue.

I warmly advice to read the original paper of the model or loss function that you are planning to use, before moving in the experimental part.