Releases: dattalab/keypoint-moseq
Keypoint MoSeq 0.4.7
Pin etils and scipy versions
Keypoint MoSeq 0.4.6
What's Changed
- fix: fixing running kruskal wallis and plotting sig syllables by @versey-sherry in #137
- Analysis fix by @versey-sherry in #144
- merge dev into main by @versey-sherry in #145
- Better error catching + change default iters for apply_model by @calebweinreb in #146
Full Changelog: 0.4.5...0.4.6
Keypoint MoSeq 0.4.5
Added FreiPose loader and function for saving keypoints.
Loading keypoints from FreiPose
To load keypoints output by FreiPose, use:
coordinates, confidences, bodyparts = kpms.load_keypoints(filepath_pattern, "freipose")
Note that confidences
is set uniformly to 1, since the FreiPose scores range from <0 to >1 and therefore can't be used for modeling. Also bodyparts
is None because these are not stored in the FreiPose output.
Initializing a config from FreiPose
Its also possible to setup a kpms config using a FreiPose config:
kpms.setup_project('test', freipose_config="path/to/skeleton_config.cfg.json")
The FreiPose config will be used to populate "bodyparts", "use_bodyparts" and "skeleton" in the kpms config. In some cases, the FreiPose skeleton may use a pair of keypoints to define one end of a limb. In FreiPose, this is used to indicate a limb midpoint. Since the same option is not available in kpms, the first bodypart in the pair is selected as the end of the limb.
Export keypoints
A new convenience function has been added for exporting keypoints:
kpms.save_keypoints(save_dir, coordinates, confidences=confidences, bodyparts=bodyparts)
One csv file is saved for each recording in coordinates
. Each row in the csv corresponds to one frame and the columns are named
"BODYPART1_x", "BODYPART1_y", "BODYPART1_conf", "BODYPART2_x", ...
Columns with confidence scores are ommitted if confidences
is not provided. Besides confidences, there can be 2 or 3 columns for each bodypart, depending on whether the keypoints are 2D or 3D.
Keypoint MoSeq 0.4.4
Added support for loading DLC files with unique bodyparts
Keypoint MoSeq 0.4.3
Fixed bug in calibration widget
Keypoint MoSeq 0.4.2
Bugfixes
- Remove "linestyle" keyword from analysis code
- Pad states before unbatch rather than after - prevents error for sequences with length < nlags
Keypoint MoSeq 0.4.1
Bumped jax-moseq dependency to version 0.2.1
Keypoint MoSeq 0.4.0
Important note!
Because of the change in jax-moseq version (see below), models trained before this update will no longer be compatible with the modeling code. Therefore, if you plan to update and have previous keypoint-MoSeq projects, we recommend installing the new code version in a separate conda environment so that you can still interact with existing projects.
Changes
-
Updated jax-moseq dependency to version 0.2.0
-
Added option for location-aware modeling (see docs for details)
-
Added a new section docs section and associated code for model comparison, selection and averaging
-
Added option to load data from facemap
-
Updated kappa scan example code to include both autoregressive-only and full-model modeling steps
-
Fixed conda environment file for Windows GPU install (thanks @!)
-
Added support for 3D data in statistical analysis pipeline (thanks @)
-
Bugfixes
Keypoint MoSeq 0.3.2
Another analysis pipeline bugfix related to seaborn version update
Keypoint MoSeq 0.3.1
Bug fix in analysis pipeline