Skip to content

Latest commit

 

History

History
21 lines (17 loc) · 597 Bytes

README.md

File metadata and controls

21 lines (17 loc) · 597 Bytes

Active Learning for Single Class Malignant Tumor Detection in Mammograms

Dataset Preparation

  • Obtain the AIIMS data from sumanyu in a particular format.
  • Make a data directory in the root of this project.
  • Put the AIIMS data in this data directory.

Scripts Description

  • aiims_eval.py: For finding fROC on AIIMS test data.
  • train.py: For training on some percentage of training data.
  • AL.py: For Active Learning iterations.

Dependencies

  • Anaconda3
  • Python>=3.6

Environment

  • Create a conda env from .yml specified.
conda create env -f env.yml