punctadetection
Puncta Detection using active learning
Science Score: 26.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (11.1%) to scientific vocabulary
Repository
Puncta Detection using active learning
Basic Info
- Host: GitHub
- Owner: dahliajones24
- License: apache-2.0
- Language: Python
- Default Branch: main
- Size: 171 MB
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
Plug and Play Active Learning for Puncta Detection
The implementation of our paper can be found: Active learning for puncta image detection in fluorescence microscopy (Machine Learning for Biomedical image analysis)
Requirements
- The codebase is built on top of MMDetection, which can be installed following the offcial instuctions.
Usage
Installation
shell
python setup.py install
Setup dataset
- Place your dataset as the following structure (Only vital files are shown). It should be easy because it's the default MMDetection data placement)
PPAL | `-- data_puncta | |--puncta | |--train |--val `--annotations | |--instances_train.json `--instances_val.json - Install datasets into datapuncta/datasetup.
- Please download groundtruth.
Please download rgb_images.
- Set up active learning datasets
shell python data_processor.py
- Set up active learning datasets
- Set up active learning datasets
shell python instances_creation.py
- Set up active learning datasets
Set up active learning datasets
shell python al_setup.pyThe above commands will set up the Puncta Datasets. The commands will also generate three different active learning initial annotations , where the COCO initial sets contain 2% of the original annotated images, and the Pascal VOC initial sets contains 5% of the original annotated images.
The resulted file structure is as following ``
PPAL |-- datapuncta | |--puncta | | | |--train | |--val | `--annotations | | | |--instancestrain.json |--instances_val.json |--data_setup | | | |--rgb_images | |--groundtruth | |--al_setup.py | |--puncta_data_processor.py | |--instances_creation.py |--activelearning |--puncta600labeled1.json |--puncta600unlabeled1.json |--puncta600labeled2.json |--puncta600unlabeled2.json |--puncta600labeled3.json |--puncta600unlabeled_3.json
``` - Please refer to data_setup.sh and createaldataset.py to generate you own active learning annotation.
Run active learning
- You can run active learning using a single command with a config file. For example, you can run puncta experiments by
shell python tools/run_al_coco.py --config al_configs/puncta/ppal_retinanet_puncta.py --model retinanet - Please check the config file to set up the data paths and environment settings before running the experiments.
Owner
- Name: Dahlia Jones
- Login: dahliajones24
- Kind: user
- Repositories: 1
- Profile: https://github.com/dahliajones24
GitHub Events
Total
Last Year
Dependencies
- cython *
- numpy *
- docutils ==0.16.0
- recommonmark *
- sphinx ==4.0.2
- sphinx-copybutton *
- sphinx_markdown_tables *
- sphinx_rtd_theme ==0.5.2
- mmcv-full >=1.3.17
- cityscapesscripts *
- imagecorruptions *
- scipy *
- sklearn *
- mmcv *
- torch *
- torchvision *
- matplotlib *
- numpy *
- pycocotools *
- six *
- terminaltables *
- asynctest * test
- codecov * test
- flake8 * test
- interrogate * test
- isort ==4.3.21 test
- kwarray * test
- onnx ==1.7.0 test
- onnxruntime >=1.8.0 test
- pytest * test
- ubelt * test
- xdoctest >=0.10.0 test
- yapf * test