Skip to content

Latest commit

 

History

History
57 lines (48 loc) · 2.02 KB

README.md

File metadata and controls

57 lines (48 loc) · 2.02 KB

ISIC: Skin Lesion Analysis Towards Melanoma Detection Scoring

PyPI Docker Build Status

Automated scoring code for the ISIC Challenge.

Installation

Python

Python version >= 3.8 is required.

pip install isic-challenge-scoring

Docker

docker pull isic/isic-challenge-scoring:latest

Usage

Python

Segmentation (2016 Tasks 1 & 2B, 2017 Task 1, 2018 Tasks 1 & 2)

isic-challenge-scoring segmentation /path/to/ISIC_GroundTruth/ /path/to/ISIC_predictions/

Classification (2016 Tasks 3 & 3B, 2017 Task 3, 2018 Task 3, 2019 Tasks 1 & 2)

isic-challenge-scoring classification /path/to/ISIC_GroundTruth.csv /path/to/ISIC_prediction.csv

Docker

Since the application requires read access to files, Docker must mount them within the container; these examples use --mount to prevent nonexistent host paths from being accidentally created.

Segmentation (2016 Tasks 1 & 2B, 2017 Task 1, 2018 Tasks 1 & 2)

docker run \
  --rm \
  --mount type=bind,source="/path/to/ISIC_GroundTruth/",destination=/root/gt/,readonly \
  --mount type=bind,source="/path/to/ISIC_predictions/",destination=/root/pred/,readonly \
  isic/isic-challenge-scoring:latest \
  segmentation \
  /root/gt/ \
  /root/pred/

Classification (2016 Tasks 3 & 3B, 2017 Task 3, 2018 Task 3, 2019 Tasks 1 & 2)

docker run \
  --rm \
  --mount type=bind,source="/path/to/ISIC_GroundTruth.csv",destination=/root/gt.csv,readonly \
  --mount type=bind,source="/path/to/ISIC_prediction.csv",destination=/root/pred.csv,readonly \
  isic/isic-challenge-scoring:latest \
  classification \
  /root/gt.csv \
  /root/pred.csv