Lightweight framework for fast prototyping and training deep neural networks with PyTorch and TensorFlow
|size (curr.)||8176 kB|
|license||GNU Affero General Public License v3.0|
delira - A Backend Agnostic High Level Deep Learning Library
Copyright (C) 2020 by RWTH Aachen University
This software is dual-licensed under:
• Commercial license (please contact: firstname.lastname@example.org)
• AGPL (GNU Affero General Public License) open source license
delira is designed to work as a backend agnostic high level deep learning library. You can choose among several computation backends.
It allows you to compare different models written for different backends without rewriting them.
For this case,
delira couples the entire training and prediction logic in backend-agnostic modules to achieve identical behavior for training in all backends.
delira is designed in a very modular way so that almost everything is easily exchangeable or customizable.
A (non-comprehensive) list of the features included in
- Dataset loading
- Dataset sampling
- Augmentation (multi-threaded) including 3D images with any number of channels (based on
- A generic trainer class that implements the training process for all backends
- Training monitoring using Visdom or Tensorboard
- Model save and load functions
- Already impelemented Datasets
- Many operations and utilities for medical imaging
What about the name?
delira started as a library to enable deep learning research and fast prototyping in medical imaging (especially in radiology).
That’s also where the name comes from:
delira was an acronym for DEep Learning In RAdiology*.
To adapt many other use cases we changed the framework’s focus quite a bit, although we are still having many medical-related utilities
and are working on constantly factoring them out.
You may choose a backend from the list below. If your desired backend is not listed and you want to add it, please open an issue (it should not be hard at all) and we will guide you during the process of doing so.
|Backend||Binary Installation||Source Installation||Notes|
||Training not possible if backend is not installed separately|
||All backends will be installed.|
The best way to learn how to use is to have a look at the tutorial notebook. Example implementations for classification problems, segmentation approaches and GANs are also provided in the notebooks folder.
If you find a bug or have an idea for an improvement, please have a look at our contribution guideline.