niclib is a utility library to ease the development of neuroimaging pipelines using deep learning. This library attempts to offers a simple interface to commonly used functions and procedures when dealing with file storage, data operations, data and patch generators, network training, loss functions and evaluation metrics.
This library does't aim to be a complete deep learning framework or serve as full replacement of other deep learning or neuroimaging libraries. If more advanced functionality is required, we encourage the user to use the provided code as a starting point to implement their own needs.
niclib only requirement is Python 3.6, the library can then be installed using pip:
pip3 install https://nic.udg.edu/niclib/wheels/niclib-1.0b0-py3-none-any.whl
The full documentation as well as some examples can be found at: https://nic.udg.edu/niclib/
Any contribution is welcome if it implements functionality useful to other users.
-
If you are a contributor please follow the repository branching workflow: https://nvie.com/posts/a-successful-git-branching-model/
-
If you are not a contributor, feel free to fork the repository and perform a pull request.
To report a bug or issue: Use github issue tracker