Skip to content

ubamba98/EAD2020

Repository files navigation

EAD2020

The code accompanying our submission to the EndoCV 2020 challenge. Full paper here.

How to Train:

  1. Downloading and Preparing Dataset : Download script.sh file and run bash script.sh on terminal. This will install dependencies, download dataset and extract them to following folders:

    • EndoCV/EAD2020-Phase-II-Segmentation-VALIDATION
    • EndoCV/EAD2020-Phase-II_HOLDOUT
    • EndoCV/ead2020_semantic_segmentation_TRAIN
  2. Clone this repository using command git clone https://github.com/ubamba98/EAD2020.git

  3. Run Sample.ipynb notebook to start training.

  4. Trained model weights and logs will get saved in models and logs directory respectively.