Skip to content

Latest commit

 

History

History
30 lines (21 loc) · 1.26 KB

README.md

File metadata and controls

30 lines (21 loc) · 1.26 KB
Raw

Humpback Whale Identification - Kaggle Winning Solution #7

Fork of https://github.com/ducha-aiki/whale-identification-2018

Heavily based on https://github.com/radekosmulski/whale

To build (and publish) Docker image

./build.sh
# ./publish.sh  # Publish to Dockerhub (requires authentication)

Download the training data

  1. Clone this repository. cd into data. Download competition data by running kaggle competitions download -c humpback-whale-identification. You might need to agree to competition rules on competition website if you get a 403.
  2. Create the train directory and extract files via running mkdir train && unzip train.zip -d train
  3. Do the same for test: mkdir test && unzip test.zip -d test
  4. Go back to top-level directory cd ../
  5. Extract boxes python apply_bboxes.py

To run with Docker

docker pull wildme/kaggle7:latest

# Map the local ./data folder into the /data/ folder inside the container (which is symlinked from /opt/whale/data/)
NV_GPU=1,3 nvidia-docker container run -it --rm --name kaggle7 -v $(pwd)/data/:/data/ --ipc=host wildme/kaggle7:latest
NV_GPU=1,3 nvidia-docker container run -it --rm --name kaggle7 -v $(pwd)/data/:/data/ --ipc=host --entrypoint="/bin/bash" wildme/kaggle7:latest