hughxx / tsf-new-paper-taste

A code implementation of new papers in the time series forecasting field.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

tsf-new-paper-taste License

A code implementation of new papers in the time series forecasting field.

Installation | Usage

One can just clone or download the project, then run the run.py.

The requirements that needed is very common, if necessary, one can install it by self.

Implemented Model

PatchMixer: https://arxiv.org/abs/2310.00655

SegRNN: https://arxiv.org/abs/2308.11200

iTransformer: https://arxiv.org/abs/2310.06625

TSMixer: https://arxiv.org/abs/2303.06053

Result

One can view result at below link, and add comment.

https://docs.google.com/spreadsheets/d/1_8WsqhCjRtVgLnGvE5VuGxUOPYzVuE9gog32o3I2LQA/edit?usp=sharing

Others

1. The model implementation strives to be as consistent as possible with the paper, but there is no guarantee of complete fidelity.

2. So far, some performance has not reached the level described in the paper, Welcome discuss and collaborate to improve it!

About

A code implementation of new papers in the time series forecasting field.


Languages

Language:Python 99.0%Language:Shell 1.0%