Metadata-Version: 2.1
Name: cvpl-tools
Version: 0.8.dev1
Summary: A Python package for utilities and classes related to the file I/O, dataset record keeping and visualization for image processing and computer vision.
Home-page: https://github.com/khanlab/cvpl_tools
License: MIT
Author: Karl5766
Author-email: karl29029@gmail.com
Requires-Python: >=3.9
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Requires-Dist: dask (>=2024.2.0)
Requires-Dist: fsspec (>=2024.6.1)
Requires-Dist: matplotlib (>=3.9)
Requires-Dist: napari (>=0.4.19)
Requires-Dist: nibabel (>=5.2.1)
Requires-Dist: numpy (>=1.23)
Requires-Dist: ome-zarr (>=0.9.0)
Requires-Dist: pillow (>=7.1.0)
Requires-Dist: scikit-image (>=0.22.0)
Requires-Dist: scipy (>=1.12.0)
Requires-Dist: zarr (>=2.17.0)
Project-URL: Repository, https://github.com/khanlab/cvpl_tools
Description-Content-Type: text/markdown

## Computer Vision Pipeline Tools

[`cvpl_tools` documentation](https://www.khanlab.ca/cvpl_tools/index.html)

cvpl_tools is a Python package for utilities and classes related to the file I/O, dataset record keeping for image processing and computer vision.

Functionalities this repository provides for managing datasets using DatasetReference:
- DatasetReference keeps an ordered array of image file paths
- Images are assigned unique IDs used to match with their annotation files
- DatasetReference can be written to disk as json using strenc.py, recording down the image files used in a processing step and their file paths
- cvpl_tools/fs.py provides interface to turn image files to and write them from numpy arrays
- reader_from_cmd.py defines a utility function for creating DatasetReference by pattern matching image files to be included in the dataset
- cvpl_tools/im folder provides an extensible segmentation pipeline that supports Napari visualization and intermediate result caching/checkpointing functionalities

### Dependency
This repository uses [Poetry](https://python-poetry.org/docs/) to manage dependency.
