oliverguhr/transformer-time-series-prediction
A proof-of-concept PyTorch implementation of transformer models for single-step and multi-step time series forecasting.
Not currently ranked — collecting fresh signals.
star history
This repository provides two transformer models for time series prediction tasks. The single-step model predicts the next time point given a sequence, while the multi-step model generates longer sequences by autoregressive prediction. Both implementations use PyTorch and were trained on temperature datasets as demonstration. The author explicitly recommends production-grade alternatives like flow-forecast for real-world use cases.
Frequently asked
- What is oliverguhr/transformer-time-series-prediction?
- A proof-of-concept PyTorch implementation of transformer models for single-step and multi-step time series forecasting.
- Is transformer-time-series-prediction open source?
- Yes — oliverguhr/transformer-time-series-prediction is open source, released under the MIT license.
- What language is transformer-time-series-prediction written in?
- oliverguhr/transformer-time-series-prediction is primarily written in Python.
- How popular is transformer-time-series-prediction?
- oliverguhr/transformer-time-series-prediction has 1.5k stars on GitHub.
- Where can I find transformer-time-series-prediction?
- oliverguhr/transformer-time-series-prediction is on GitHub at https://github.com/oliverguhr/transformer-time-series-prediction.