SalesforceAIResearch/uni2ts
Unified Training of Universal Time Series Forecasting Transformers
This library helps data scientists and machine learning engineers accurately predict future trends using time series data. You input historical data, and it outputs predictions for what will happen next. It's designed for anyone needing to forecast quantities over time, such as sales, stock prices, or sensor readings.
1,436 stars.
Use this if you need a robust, pre-trained model to make accurate predictions on diverse time series datasets without extensive custom model development.
Not ideal if you're a business user looking for a point-and-click solution or don't have experience with Python and machine learning frameworks.
Stars
1,436
Forks
191
Language
Jupyter Notebook
License
Apache-2.0
Category
Last pushed
Jan 29, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/SalesforceAIResearch/uni2ts"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related models
amazon-science/chronos-forecasting
Chronos: Pretrained Models for Time Series Forecasting
moment-timeseries-foundation-model/moment
MOMENT: A Family of Open Time-series Foundation Models, ICML'24
ServiceNow/TACTiS
TACTiS-2: Better, Faster, Simpler Attentional Copulas for Multivariate Time Series, from...
yotambraun/APDTFlow
APDTFlow is a modern and extensible forecasting framework for time series data that leverages...
Thinklab-SJTU/Crossformer
Official implementation of our ICLR 2023 paper "Crossformer: Transformer Utilizing...