trainer
https://github.com/kubeflow/trainer
Go
Distributed AI Model Training and LLM Fine-Tuning on Kubernetes
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Go not yet supported1 Subscribers
Add a CodeTriage badge to trainer
Help out
- Issues
- TrainJobStatus silently overwrites conflicting runtime volume or volumeMount
- chore(docs): add gRPC API schema for SearchAlgorithmService to KEP-3562
- fix(runtimes): mount Flux shared-memory volume as writable
- Decouple UTs between Framework and Plugins packages
- Automate managing of network policy resources for every train job
- Support for Tensor Caching in Kubeflow Data Cache
- fix(api): reject non-integral bounds for Int search spaces
- Transparent GPU Checkpointing of TrainJobs with CRIU
- Support Multi-Node NVLink (MNNVL) for TrainJob with Topology Aware Scheduling
- Ship a default Grafana dashboard for Kubeflow Trainer
- Docs
- Go not yet supported