Model description

Following the new support for Time Series Transformers in the API (and the great blog by @NielsRogge and @kashif here), I propose adding "Informer" - AAAI 2021 Best Paper model.

• Paper: Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
  
• Model implementation: https://github.com/zhouhaoyi/Informer2020

Why this model?

Compared to other forcasting transformers (see below), Informer seems to be the most "code-stable" one, with the most starts & forks in github.

Popular forcasting Transformes, with link to the repository:

LogTrans - NIPS 2019
Informer - AAAI 2021 (Best Paper)
Autoformer - NIPS 2021
Pyraformer - ICLR 2022
FEDformer - ICML 2022

This list based on the paper: Are Transformers Effective for Time Series Forecasting? (AAAI-23)

I would like to implement the model :)

Thank you,
Eli

Open source status

• The model implementation is available
• The model weights are available

Provide useful links for the implementation

@zhouhaoyi - repository creator