Skip to content

Latest commit

 

History

History
21 lines (15 loc) · 1.61 KB

File metadata and controls

21 lines (15 loc) · 1.61 KB

training_ML_for_image_analysis_EBI

Group Project: "Machine Learning for Image Analysis" as a part of Computational Biology track of EBI Predoc Course 2019

In this project we will write a Machine Learning pipeline for segmenting nuclei in microscopy images from Kaggle Data Science Bowl competition. Participants will write their own data loader, model, loss and training loop.

The training is written for Google Colab Notebooks.

Here are some examples of the images we will be working with:

alt text alt text alt text alt text

Useful links

  1. Machine Learning course by Andrew Ng on Coursera
  2. Convolutional Neural Networks for Visual Recognition - a Stanford course
  3. PyTorch Tutorials - Getting Started Tutorials for Image and Text processing
  4. UNet - The Intuition Behind UNet
  5. Focal and Dice Loss - Investigating Focal and Dice Loss for the Kaggle 2018 Data Science Bowl