Granite Timeseries Patchtst
Part of the IBM Granite time series foundation model family, this PatchTST model was pretrained on the ETTh1 dataset for time series prediction. It uses a transformer architecture with patching to capture long-range dependencies in multivariate data. The model is available on Hugging Face and integrates with the Transformers library for forecasting tasks.
Open SourceApache-2.0WebCLIAPI
876stars
278forks
3features
2023since
Overview
3 featuresIn the Data science & ML workbench space, Granite Timeseries Patchtst takes a focused approach. Accurate long-term forecasting of multivariate time series data using transformer-based models. It is built as an open-source project for data scientists. The project is open source (Apache-2.0). It runs on the web, the command line, and API.
IBM Granite builds and maintains Granite Timeseries Patchtst, and it first shipped in 2023. Development happens publicly on GitHub with 876 stars and 59 commits in the last 90 days. Among its 3 catalogued features are time series forecasting, patchTST architecture, and pretrained weights.
Summary written by a language model from the project’s public pages.
- Time series forecasting
- PatchTST architecture
- Pretrained weights
Tags
time-seriesforecastingpatchtstgraniteibm
AI capabilities
Structured
Inference: LocalWeights: Open
Built with & integrations
Hosting
AWS
Runs on
BrowserCLIAPI-only
Detected from
- AWS
- x-amz-cf-id header · x-amz-cf-pop header · via header
Trust & compliance
License
Apache-2.0
Verified signals
Indexing history
1What PulseGate has recorded for this listing
- Indexed1 Aug · 13:21 UTCibm-granite/granite-timeseries-patchtst verified against its public sourceSource: PulseGate · Open
Frequently asked questions about Granite Timeseries Patchtst
- What does Granite Timeseries Patchtst do?
- Accurate long-term forecasting of multivariate time series data using transformer-based models. It is catalogued under Data science & ML workbench on PulseGate.
- Who is Granite Timeseries Patchtst for?
- Granite Timeseries Patchtst is an open-source project built for data scientists.
- Is Granite Timeseries Patchtst free?
- Yes — Granite Timeseries Patchtst is open source under the Apache-2.0 license and free to use.
- What platforms does Granite Timeseries Patchtst run on?
- Granite Timeseries Patchtst runs on the web, the command line, and API.
- Is Granite Timeseries Patchtst still maintained?
- The GitHub repository shows 59 commits in the last 90 days.
- What are alternatives to Granite Timeseries Patchtst?
- Similar projects tracked by PulseGate include corvus-ai, MFGLib documentation, and probe-research.corvus-aiMFGLib documentationprobe-research
- Who makes Granite Timeseries Patchtst?
- Granite Timeseries Patchtst is developed by IBM Granite.
- When did Granite Timeseries Patchtst launch?
- Granite Timeseries Patchtst first shipped in 2023.
Also in Data science & ML workbench
Same category — not a similarity match