This is first release of VisCy, the machine learning pipeline to train and deploy computer vision models for single-cell phenotyping.
With 0.1.0 the following key features are available:
- Training, evaluation, inference, and deployment of virtual staining models based on 2D Residual U-Net, 2.5D U-Net, 3D U-Net, and UNeXt2 architectures
- Data module implementations for HCS OME-Zarr datasets, as well public test datasets like LiveCell and CTMC v1.
- Composing datasets and transformations for training and validation
- Distributed (DDP) training
The weights of the virtual staining models reported in the [preprint](https://doi.org/10.1101/2024.05.31.596901) can be found in the binaries section below.
What's Changed
* Migration from microDL by ziw-liu in https://github.com/mehta-lab/VisCy/pull/1
* Update README by ziw-liu in https://github.com/mehta-lab/VisCy/pull/22
* Bump iohub version by ziw-liu in https://github.com/mehta-lab/VisCy/pull/25
* readme + dependencies tested with python 3.10 by mattersoflight in https://github.com/mehta-lab/VisCy/pull/30
* Demo notebooks by ziw-liu in https://github.com/mehta-lab/VisCy/pull/29
* 3D augmentation and 2.1D U-Net by ziw-liu in https://github.com/mehta-lab/VisCy/pull/27
* Fix datamodule by ziw-liu in https://github.com/mehta-lab/VisCy/pull/28
* Improve augmentation by ziw-liu in https://github.com/mehta-lab/VisCy/pull/31
* Fix inference by ziw-liu in https://github.com/mehta-lab/VisCy/pull/32
* viscy -> VisCy by mattersoflight in https://github.com/mehta-lab/VisCy/pull/34
* Update readme.md typo for `cd VisCy` by edyoshikun in https://github.com/mehta-lab/VisCy/pull/41
* 2.1D upscale decoder by ziw-liu in https://github.com/mehta-lab/VisCy/pull/37
* dlmbl 2023 archive by mattersoflight in https://github.com/mehta-lab/VisCy/pull/44
* Fix center slice metrics for 3D output by ziw-liu in https://github.com/mehta-lab/VisCy/pull/51
* Configure the number of image samples logged at each epoch and batch by ziw-liu in https://github.com/mehta-lab/VisCy/pull/49
* Example workflow by ziw-liu in https://github.com/mehta-lab/VisCy/pull/45
* Project icon by ziw-liu in https://github.com/mehta-lab/VisCy/pull/38
* Fix predicting new channels in an existing store by ziw-liu in https://github.com/mehta-lab/VisCy/pull/57
* Document data methods by ziw-liu in https://github.com/mehta-lab/VisCy/pull/50
* Visualize feature maps by ziw-liu in https://github.com/mehta-lab/VisCy/pull/53
* Baseline 3D-LUNeXt by ziw-liu in https://github.com/mehta-lab/VisCy/pull/58
* Preprocess CLI and source scaling during prediction by ziw-liu in https://github.com/mehta-lab/VisCy/pull/59
* Configurable augmentations by ziw-liu in https://github.com/mehta-lab/VisCy/pull/61
* Bump dependencies and update documentation by ziw-liu in https://github.com/mehta-lab/VisCy/pull/64
* checkpoint as a model config parameter for warmup cosine learning rates by edyoshikun in https://github.com/mehta-lab/VisCy/pull/66
* Masked autoencoder pre-training for virtual staining models by ziw-liu in https://github.com/mehta-lab/VisCy/pull/67
* Filter empty detections in labels by ziw-liu in https://github.com/mehta-lab/VisCy/pull/74
* Add CITATION.cff by ziw-liu in https://github.com/mehta-lab/VisCy/pull/79
* Fix 3D to 2D prediction with UNeXt2 model by ziw-liu in https://github.com/mehta-lab/VisCy/pull/80
* 2D FCMAE by ziw-liu in https://github.com/mehta-lab/VisCy/pull/71
* Rename UNeXt2 by ziw-liu in https://github.com/mehta-lab/VisCy/pull/84
* Test on Python 3.12 by ziw-liu in https://github.com/mehta-lab/VisCy/pull/88
* Add preprint reference to README by ziw-liu in https://github.com/mehta-lab/VisCy/pull/85
* Fix architecture name in network diagram script by ziw-liu in https://github.com/mehta-lab/VisCy/pull/86
* Add the scale metadata to the output_stores by edyoshikun in https://github.com/mehta-lab/VisCy/pull/89
* bumping to cellpose 3 by edyoshikun in https://github.com/mehta-lab/VisCy/pull/92
* Scale metadata handling for positions by edyoshikun in https://github.com/mehta-lab/VisCy/pull/93
* Demo for VSCyto2D and VSCyto3D by edyoshikun in https://github.com/mehta-lab/VisCy/pull/94
* Fix demos on other platforms by ziw-liu in https://github.com/mehta-lab/VisCy/pull/95
**Full Changelog**: https://github.com/mehta-lab/VisCy/commits/v0.1.0