Deeplabcut

Latest version: v2.3.10

Safety actively analyzes 638845 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 4 of 10

2.2.0.3

to use: `pip install 'deeplabcut[gui]'==2.2.0.3`

Improvements:

* substantial GUI speed improvement -- https://github.com/DeepLabCut/DeepLabCut/commit/53ae5f2fab7dff9969ac515ab620d36615ed22b4
* allow single point tracking in maDLC https://github.com/DeepLabCut/DeepLabCut/commit/50489743b27b0a36747a1911a4cce1971dc0107c
* speed up maDLC when there is only one animal https://github.com/DeepLabCut/DeepLabCut/commit/b56691c626123116bae7b2dfbbb91af8a5e5050f
* Ability to modify fixed fps with keypoint only option (1534)
* Ability to plotting assemblies in evaluation (1560)
* Remove references to convert_raw_tracks_to_h5 and update docs (1553)
* Ability to consider symmetric keypoints when computing OKS (1551), Ability to compute mAP over train and test splits (1513)
* expanded test-suite (unit-tests) (various commits)
* updated docs in many regards (incl. Tips on video reencoding and preprocessing (1517); Add Dockerfiles and helper scripts (1511); Update print statements for video analysis, evaluation, tracking (1523))
* Typos https://github.com/DeepLabCut/DeepLabCut/commit/2d00edb583fe530580b21640cf086c4524b77ffc

Bugfixes:
* Fix KMeans segfault by jeylau in https://github.com/DeepLabCut/DeepLabCut/pull/1554
* Fix covariance matrix indexing by jeylau in https://github.com/DeepLabCut/DeepLabCut/pull/1565
* fix typo in compute_deviations() by quantumdot in https://github.com/DeepLabCut/DeepLabCut/pull/1582
* Fix effnet offsets by jeylau in https://github.com/DeepLabCut/DeepLabCut/pull/1581
* Store best edges only when they exist by jeylau in https://github.com/DeepLabCut/DeepLabCut/pull/1471
* Fix ValueError when selecting best skeleton by jeylau in https://github.com/DeepLabCut/DeepLabCut/pull/1495
* Fix PAF indexing by jeylau in https://github.com/DeepLabCut/DeepLabCut/pull/1506


Contributors
Contributors include jeylau, MMathisLab MitchFuchs, quantumdot, ehsainit, masahito1997, KonradDanielewski , stes, AlexEMG -- in particular:
* aiporre made their first contribution in https://github.com/DeepLabCut/DeepLabCut/pull/1466
* mfkeles made their first contribution in https://github.com/DeepLabCut/DeepLabCut/pull/1472
* remi-pr made their first contribution in https://github.com/DeepLabCut/DeepLabCut/pull/1505
* stes made their first contribution in https://github.com/DeepLabCut/DeepLabCut/pull/1511
* masahito1997 made their first contribution in https://github.com/DeepLabCut/DeepLabCut/pull/1530
* MitchFuchs made their first contribution in https://github.com/DeepLabCut/DeepLabCut/pull/1580
* quantumdot made their first contribution in https://github.com/DeepLabCut/DeepLabCut/pull/1582

Funding includes [ChanZuckerberg Initiative's Essential Open Source Software for Science](https://chanzuckerberg.com/eoss/). Thanks!

2.2

AUTHORS: https://github.com/DeepLabCut/DeepLabCut/blob/master/AUTHORS

2.2.0

This is a **MAJOR** change to the stable release!

This release comes following 3 pre-releases, all documented (see links below), which we highly encourage you to read. There are major changes since maDLC was beta released in 2020, with massive performance gains. We recommend you use the new conda file, and re-train your older labeled data for best performance. Note, we include a [new COLAB](https://github.com/DeepLabCut/DeepLabCut/blob/master/examples/COLAB_maDLC_TrainNetwork_VideoAnalysis.ipynb) notebook that we highly recommend you using and/or minimally reviewing for your local use.

2.2rc3

To upgrade: `pip install 'deeplabcut[gui]'==2.2rc3`

We have updated and refactored major parts of the code base for seamless Tensorflow 2+ integration. This means:

(1) You should update: you can create a new installation from our new conda environment. This is easy, and you can keep your older env for safe keeping! In short, your older projects will work, and new projects will get all the cool advantages of TF2!), but in short its quite a bit easier to build on, so we hope this enables more researchers to play with the base code.

How do I easily update?

Simple: download the NEW conda environment file. You can keep your older DLC-GPU, DLC-CPU for safe keeping. Then, as this is a release candidate, just run: `pip install 'deeplabcut[gui]'==2.2rc3` in this new environment.

(2) deeplabcut-core is depreciated! ✅

(3) The latest NVIDIA GPUs, CUDA, etc are supported! 🥂🙌

(4) We tested it, a lot …. a big shout out to lead developer Dr. Jessy Lauer for the big PR and testing the code across platforms, GPUs, models and TF versions to be sure we did not slow you down! See full PR here: https://github.com/DeepLabCut/DeepLabCut/pull/1323 and here are some take homes:

Benchmarked on 4 datasets (single- and multi-animal, w/ grayscale and color images) with TensorFlow (TF)1.15.5 (which serves as reference), TF2.3, and TF2.5; batch size 8, 30k iterations (except for the marmosets: 20k); 3 backbones (resnet_50, mobilenet_v2_0.5, efficientnet-b0); 2 GPU devices (TITAN RTX & GEFORCE GTX 1080). No significant main effects of either backbone or TF version were found. Training duration is reported relative to TF1 training time (Y axis, and value printed above each bar), and in seconds (underneath/within the bar):

![123963082-682b3400-d9b2-11eb-9c3e-16ccf45a7589](https://user-images.githubusercontent.com/28102185/124033500-4ce41700-d9fa-11eb-854b-d2674bf07b55.png)

(5) We have new docs to help you with the transition. This is simpler to install in the long run (1 conda file!) and again just requires you have CUDA (and associated cuDNN, see docs!).

Along with this major change there are some excellent updates to the code base for RC3!

For the full change log, see here: https://github.com/DeepLabCut/DeepLabCut/compare/2.2rc2...master

Other highlights include:

- compatibility with matplotlib changes

- New projects created with Japanese character compatibility (thanks bobfromjapan !)

- Support images that are smaller (both sides or one side) than the desired crop size (1318), thanks lab member Shaokai Ye aka yeshaokai !

- Optionally ignore some body parts during tracking (1348) —> this is ideal in maDLC if you have an animal where most body parts are compact, but some far (i.e., the tail of the mouse). You can leverage this!

- better error messages, small 🐛 squashing, and other housekeeping things. Including one regarding destfolders found by new lab member Aristotelis Economides intergalactic-mammoth

2.2rc2

to use: `pip install 'deeplabcut[gui]'==2.2rc2`

**New features:**
- API update to create video with all detections (easier to use!): `deeplabcut.create_video_with_all_detections(config_path, ['/fullpath/project/videos/testVideo.mp4'])` --> now you can make a video right after pose estimation only step quickly. https://github.com/DeepLabCut/DeepLabCut/pull/1250
- DLCRNet new default net!🔥 https://github.com/DeepLabCut/DeepLabCut/pull/1255
- you can fund us in 1 click 🙏🤩 https://github.com/DeepLabCut/DeepLabCut/commit/487154d2a15758a14642b521a1da0413a0e6a28c
💖 https://github.com/sponsors/DeepLabCut
- smarter inclusion of identity information: https://github.com/DeepLabCut/DeepLabCut/pull/1251
- [x] Fix missing identity information in tracklets
- [x] Allow animal assembly with identity only
- [x] Automatically stitch with identity when possible
- [x] Allow tracking with identity only (optimal ID assignment based on soft voting of body part identity predictions)
use: `deeplabcut.convert_detections2tracklets(..., identity_only=True)`
read more [here!](https://github.com/DeepLabCut/DeepLabCut/blob/master/docs/maDLC_UserGuide.md#optimized-animal-assembly--video-analysis)
- Check for the addition of new animals to the config while labeling (namely, if you find you have MORE animals that you thought in your maDLC project, just add a new name to the config.yaml, and then re-open labeling GUI and go!) 1273
- user can define frame rate of camera for make labeled videos: https://github.com/DeepLabCut/DeepLabCut/pull/1240
- dynamically allocate memory on GPU in tensorflow for video analysis `deeplabcut.analyze_videos(config_path, [new_video_path], allow_growth=True)` https://github.com/DeepLabCut/DeepLabCut/issues/1246


**upgrades:**
- select DLCRNet from main project manager GUI! https://github.com/DeepLabCut/DeepLabCut/commit/5a460362c1fe0d22a3475330a6216a594b8cb21e
- installation docs updated
- tracklet docs upgraded! https://github.com/DeepLabCut/DeepLabCut/pull/1239
- docstring upgrade (thanks Joilence) https://github.com/DeepLabCut/DeepLabCut/pull/1253
- Roadmap updated! https://github.com/DeepLabCut/DeepLabCut/pull/1255
- user-defined skeleton possible (but not advised per se): https://github.com/DeepLabCut/DeepLabCut/pull/1256
- much faster dataset creation! https://github.com/DeepLabCut/DeepLabCut/pull/1257
- if symlink fails, move videos: https://github.com/DeepLabCut/DeepLabCut/pull/1272


**Bug fixes:**
- single animal mode supported in new tracklet stitcher: https://github.com/DeepLabCut/DeepLabCut/pull/1219
- return 0 mAP when no reasonable assemblies are found (vs just indexingError); i.e., more informative error now: https://github.com/DeepLabCut/DeepLabCut/pull/1220
- force dataframe to start at 0 (even if not animal visible): https://github.com/DeepLabCut/DeepLabCut/pull/1225
- image panel error fixed: https://github.com/DeepLabCut/DeepLabCut/pull/1231
- path clean up (thanks sin-mike) https://github.com/DeepLabCut/DeepLabCut/commit/fd922f4ccd0be4196953afb2df55751b2700b7e8
- use the user-input individual names https://github.com/DeepLabCut/DeepLabCut/pull/1267
- config integrity on re-crop fix: https://github.com/DeepLabCut/DeepLabCut/pull/1274
- slicing error (thanks backyardbiomech!) https://github.com/DeepLabCut/DeepLabCut/commit/15d1fef96a545c6d791c412f355191139bb5bb51


Thank you to all the contributors (issues raised, code fixes, and suggestions!)!
jeylau AlexEMG MMathisLab and those tagged above.

Extra shout out to backyardbiomech and KonradDanielewski for amazing support across the community platforms and feedback to core-dev team!

2.2rc1

Page 4 of 10

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.