SLEAP v1.0.10a6

Pre-release of minor version update with performance tweaks and bug fixes.

Changelog

• Update to TensorFlow 2.1.2 (security patch)
• Switch to ID-based hashing for LabeledFrame. This dramatically increases the performance of frame manipulation operations.
• Several convenience methods for sleap.Labels:
  • Add describe method to Labels for easy inspection of dataset stats
  • Add has_frame method to Labels for quick checking of frame existence
  • Add remove_user_instances and remove_predictions for quick dataset cleanup
• Remove predicted instances in existing frames before merging in active learning results (fixes #413)
• Conda environment.yml clean-up: de-duplicates dependencies managed by pip
• Set h5py version requirement to 2.10.0 to prevent TensorFlow model loading issue
• Added experimental maDLC CSV labels importing support (#412)

Installing

Using Conda (Windows):
Create new environment sleap_alpha (recommended):
conda create -n sleap_alpha -c sleap/label/dev sleap=1.0.10a6
or to update inside an existing environment:
conda install -c sleap/label/dev sleap=1.0.10a6

Using PyPI (Linux/Mac):
pip install sleap==1.0.10a6

SLEAP v1.0.10a5

Pre-release of minor version update with performance tweaks and bug fixes.

Changes:

• Update to TensorFlow 2.1.2 (security patch)
• Switch to ID-based hashing for LabeledFrame. This dramatically increases the performance of frame manipulation operations.
• Several convenience methods for sleap.Labels:
  • Add describe method to Labels for easy inspection of dataset stats
  • Add has_frame method to Labels for quick checking of frame existence
  • Add remove_user_instances and remove_predictions for quick dataset cleanup
• Remove predicted instances in existing frames before merging in active learning results (fixes #413)
• Conda environment.yml clean-up: de-duplicates dependencies managed by pip

SLEAP v1.0.10a4

Pre-release of minor version update with performance tweaks and bug fixes.

Changes:

• Update to TensorFlow 2.1.2 (security patch)
• Switch to ID-based hashing for LabeledFrame. This dramatically increases the performance of frame manipulation operations.
• Several convenience methods for sleap.Labels:
  • Add describe method to Labels for easy inspection of dataset stats
  • Add has_frame method to Labels for quick checking of frame existence
  • Add remove_user_instances and remove_predictions for quick dataset cleanup
• Remove predicted instances in existing frames before merging in active learning results (fixes #413)
• Conda environment.yml clean-up: de-duplicates dependencies managed by pip

SLEAP v1.0.10a2

Pre-release of minor version update with performance tweaks.

Changes:

• Switch to ID-based hashing for LabeledFrame. This dramatically increases the performance of frame manipulation operations.
• Version fence numpy for compatibility with TensorFlow.

SLEAP v1.0.10a1

Pre-release of minor version update with performance tweaks.

Changes:

• Switch to ID-based hashing for LabeledFrame. This dramatically increases the performance of frame manipulation operations.

SLEAP v1.0.9

Minor version update with mostly small bug fixes. This is the last release in the 1.0.x series. SLEAP 1.1.x will contain a major overhaul of the inference module and updates to TensorFlow 2.3.

Changelog

• Hotfix for #388.
• More docstrings and coverage.
• Hotfix for #389: Disable GPU memory preallocation during training by default.
• Add PAF magnitude visualization during training for bottom-up models.
• Validate skeleton for bottom-up models in the GUI before training (#354)
• Re-use empty GUI window when opening a new project.
• Fix part affinity field preview overlay.
• Fix track serialization bug when using Kalman filter.

Installing

Using Conda (Windows):
Create new environment sleap (recommended):
conda create -n sleap -c sleap sleap=1.0.9
or to update inside an existing environment:
conda install -c sleap sleap=1.0.9

Using PyPI (Linux/Mac):
pip install sleap==1.0.9

SLEAP v1.0.9a3

• Hotfix for #389: Disable GPU memory preallocation during training by default.

SLEAP v1.0.9a2

• Hotfix for #388.
• More docstrings and coverage.

SLEAP v1.0.9a1

Automated release of 1.0.9a1 on 2020-07-30

(try 2, not using api token)

SLEAP v1.0.8

Bug fixes + tweaks to metrics GUI