Template for extending ASReview with a new dataset

ASReview has support for extensions, which enable you to seamlessly integrate your own programs with the ASReview framework. This template can extent ASReview with new data.

See the section Extensions on ReadTheDocs for more information on writing extensions.

Getting started

Click the Use this template button and add/modify the algorithms. Install your new dataset with

pip install .

or

pip install git+https://github.com/{USER_NAME}/{REPO_NAME}.git

and replace {USER_NAME} and {REPO_NAME} by your own details.

Usage

Adding a dataset to ASReview is done by extending the BaseDataSet class, adding it to a BaseDataGroup and finally, adding it to ASReview. To use this template, fork it and modify the following files:

• A BaseDataSet object and a BaseDataGroup are defined in asreviewcontrib/dataset_name/your_dataset.py. Modify this file to add your own datasets. The BaseDataSet class should always be added to a BaseDataGroup object.

• Adding your BaseDataGroup object to ASReview is done via the asreviewcontrib/dataset_name/__init__.py file. This file should import your BaseDataGroup.

• Adjust setup.py with information about your dataset, and define the dataset entrypoint by adding your BaseDataGroup.

• Add your dataset to the data folder of the template.

For advanced usage, check out the BaseDataGroup in the example and the documentation.

License

MIT license