The package module pychunklbl.toolkit provides eight parameterized functions designed to work with large image files and provide pre-processing for machine learning.
The package module is intended developers to create machine learning datasets leveraging the openslide library for usage with tensorflow 2.0.
pip install digipath_mltk
# requires python 3.5 or later
pip3 install -r requirements.txt
pip3 install -r requirements.txt
pip3 install --upgrade ../DigiPath_MLTK
Images used in the examples below was downloaded from openslide data, the other data files are in the repository data/ directory.
python3 -m digipath_mltk.cli -m write_mask_preview_set -i yourpath/images/CMU-1-Small-Region.svs -o results
python3 -m digipath_mltk.cli -m wsi_to_patches_dir -i CMU-1-Small-Region.svs -o results
python3 -m digipath_mltk.cli -m wsi_to_tfrecord -i CMU-1-Small-Region.svs -o results
python3 -m digipath_mltk.cli -m tfrecord_to_masked_thumb -i CMU-1-Small-Region.svs -r CMU-1-Small-Region.tfrecords -o results
( test data not currently available in DigiPath_MLTK repository for the following examples )
python3 -m digipath_mltk.cli -m registration_to_dir -i wsi_fixed.tiff -f wsi_float.tiff -D wsi_pair_sample_offset.csv -o results
python3 -m digipath_mltk.cli -m registration_to_tfrecord -i wsi_fixed.tiff -f wsi_float.tiff -D wsi_pair_sample_offset.csv -o results
Find the patches in a wsi file defined in an annotations file with a priority file and write to a directory.
python3 -m digipath_mltk.cli -m annotations_to_dir -i wsi_float.tiff -p wsi_float_annotation.csv -a wsi_float_annotation.xml -o results
Find the patches in a wsi file defined in an annotations file with a priority file and write to a tfrecords file.
python3 -m digipath_mltk.cli -m annotations_to_tfrecord -i wsi_float.tiff -p wsi_float_annotation.csv -a wsi_float_annotation.xml -o results