yuqinie98/PatchTST
Official implementation of PatchTST, a Transformer model that segments time series into patches for long-term forecasting.
Not currently ranked — collecting fresh signals.
star history
PatchTST is a Transformer-based architecture for long-term time series forecasting. It segments time series into subseries-level patches that serve as input tokens to the Transformer, along with channel-independence where each univariate series shares embedding weights. The model achieves significant MSE and MAE reductions over prior Transformer-based and non-Transformer baselines. It has been adopted into GluonTS, NeuralForecast, and tsai libraries.
Frequently asked
- What is yuqinie98/PatchTST?
- Official implementation of PatchTST, a Transformer model that segments time series into patches for long-term forecasting.
- Is PatchTST open source?
- Yes — yuqinie98/PatchTST is open source, released under the Apache-2.0 license.
- What language is PatchTST written in?
- yuqinie98/PatchTST is primarily written in Python.
- How popular is PatchTST?
- yuqinie98/PatchTST has 2.6k stars on GitHub.
- Where can I find PatchTST?
- yuqinie98/PatchTST is on GitHub at https://github.com/yuqinie98/PatchTST.