emfdavid

emfdavid/insitubatch

Train in place on n-dimensional cloud tensors.

6 good first / help-wanted issues · Python · last activity Aug 16, 2026

13 stars 2 forks 13 watchers Python MIT License

View on GitHub Website

6 Open Issues Need Help Last updated: Aug 16, 2026

Open Issues Need Help

View All on GitHub

Vectorize the per-sample random crop in `examples/wb2_dataloader.py` 2 months ago

enhancement good first issue

emfdavid/insitubatch

13

Train in place on n-dimensional cloud tensors.

Python

Free-threaded bench: report p50/p95 instead of median and min–max 2 months ago

enhancement help wanted

emfdavid/insitubatch

13

Train in place on n-dimensional cloud tensors.

Python

`window_factor` sizes residency by offset span, not by the offset set 2 months ago

bug help wanted performance

emfdavid/insitubatch

13

Train in place on n-dimensional cloud tensors.

Python

Add `samples_for_time_range()` — a no-xarray coordinate helper for `sample_range` 2 months ago

enhancement good first issue

emfdavid/insitubatch

13

Train in place on n-dimensional cloud tensors.

Python

Expose a `dropped_edge_anchors` count on `InSituDataset` 2 months ago

enhancement good first issue

emfdavid/insitubatch

13

Train in place on n-dimensional cloud tensors.

Python

examples/README.md: wrong class name, and "cross-domain companion" is now one of three 2 months ago

documentation good first issue

emfdavid/insitubatch

13

Train in place on n-dimensional cloud tensors.

Python