Sleap

Latest version: v1.4.0

Safety actively analyzes 641171 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 3 of 19

1.2.5

See the [Installation page](https://sleap.ai/installation.html) in the docs for more info.

Highlights
Add read/write adaptor for ndx-pose (835, 845)

Full changelog
Documentation
Update assisted-labeling docs, installation docs, and environment_no_cuda.yml (847)

Enhancements
Add read/write adaptor for ndx-pose (835, 845)
Expose attributes of NWBFile and create Labels API for exporting to NWB (855)

Fixes
Change existing skeleton to match loaded skeleton (840)
Recalculate crop size if user-specified crop size indivisible by max stride (841)

1.2.4

See the [Installation page](https://sleap.ai/installation.html) in the docs for more info.

Highlights
- Add option to predict on all videos (749)
- Add dropdown to choose video to generate suggestions on (786)
- Add button to toggle grayscale of current video (788)

Full changelog
Documentation
- Add links to discussion (748)
- Contributing Guide, Code of Conduct, and Issues Template (746)
- Update tracking docs (761)

Enhancements
- Add option to predict on all videos (749)
- Create multiple analysis files for multi-video projects (768)
- Add button to toggle grayscale of current video (788)
- Choose video to generate suggestions (786)
- Add CLI sleap-render command to render videos (796)
- Allow user to set grayscale when replacing videos (mp4/avi only) (787)
- Support grayscale for SingleImageVideo backend (789)

Fixes
- Generate suggestions for videos with less frames than samples per video (781) (783)
- Fix h5py dependency (815)
- Remove low-scoring predictions before merging inference results (817)

1.2.3

See the [Installation page](https://sleap.ai/installation.html) in the docs for more info.

Highlights
- Add support for importing [AlphaTracker](https://github.com/ZexinChen/AlphaTracker) annotations
- Add support for new DeepLabCut labels formats

Full changelog
Enhancements
- Add edges to analysis h5 (707)
- Speed-up labeling suggestions look-up (709)

Fixes
- Add support for new single animal DLC format (704)
- Update links from murthylab to talmolab (724)
- Pinned the pip conda package to `conda-forge::pip<=22.0.3` to fix hanging issues (724, 726)

1.2.2

See the [Installation page](https://sleap.ai/installation.html) in the docs for more info.

Highlights
- Major inference speed improvements of 2-4x when using the high-level API:
_Old:_
![inference old](https://user-images.githubusercontent.com/3187454/161452502-d280d7b9-f4bf-46bf-a895-4f286bfcdb8f.gif)
_New:_
![inference new freeze](https://user-images.githubusercontent.com/3187454/161452508-c56d8d4c-0794-4738-b0c5-1f55750fa3ac.gif)

- New training monitor statistics and more detailed graphics in the loss plot:
![image](https://user-images.githubusercontent.com/3187454/161452878-c122e044-4df2-4f4e-bffe-3509a14ccd4a.png)


Full changelog
Enhancements
- Add support for new maDLC labels format (678)
- Training monitor enhancements (implements 624) (691)
- Add hide instance menu item and hotkey (<kbd>H</kbd>) (implements 665) (692, 694)

Fixes
- Fix numpy conversion in inference (687)
- Fix topological sorting to always start with root node (688)
- Create unique default shortcuts (fixes 686) (690)
- Fix formatting to read and save tracking scores (693)

1.2.1

See the [Installation page](https://sleap.ai/installation.html) in the docs for more info.

Full changelog
- Add support for new maDLC labels format (678)
- Fixes 676

- Enable TensorFlow 2.8 support (683)
- Fixes Colab support (680)

- Add experimental support for supervised ID models (679)
- **Note:** This is being released for backward compatibility, but it should be considered strictly experimental. It is not yet available in the GUI. Full functionality will be documented in a future release.

1.2.0

Page 3 of 19

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.