This Python module contains tools and templated code to facilitate the training and validation of deep learning algorithms for medical imaging applications. Links to tutorials are available at the end of this README.md.
The code and tutorials in this repository work without any additional configuration if you load via links to Google Colob below. If you wish to replicate this environment on a local machine of Jupyter server, you will need to first install the dl_utils library. More information can be found here:
https://github.com/peterchang77/dl_utils
Any *.ipynb tutorial can be run directly from GitHub in a Google Colab hosted Jupyter instance using the following URL pattern:
https://https://colab.research.google.com/github/peterchang77/dl_core/blob/master/docs/notebooks/[name-of-folder]/[name-of-notebook.ipynb]
Here is an overview of the currently available notebooks with direct Google Colab links:
- Introduction to Tensorflow / Keras 2.0 API Google Colab
- How to Create a Data Client Google Colab
- Overview of U-Net Google Colab
- Overview of U-Net and Custom Loss / Metrics Google Colab
- Overview of Regression Network Google Colab