Clue-WATTx hackathon is my contribution to the hackathon.
- Website: http://cluehackathon.wattx.io/
- Original example submission repository: https://github.com/WATTx/cluehackathon
The submission and starting point of train scripts are copied from the above repository.
My idea is to collaborate towards a better result at the end of the hackathon.
Everyone is invited to use the code, w/o my consent and knowledge, and use it the way they think is appropriate to the task at hand. Needless to say, pull requests, branches, and ideas are more than welcome.
In any chance, my submissions from this repository win any prize, everyone with any accepted pull-request will have an equal share of the prize. Ideally, everyone basing any submission on this repo would do the same, but they don't have to, and I claim nothing out of their changes and submissions.
IMPORTANT: Do NOT add the sample or the synthesized data to this repository or any clones.
The data is the property of the Clue company (BioWink GmbH), and I believe we do not have
the permission to publish it publicly. It is safe to copy the data under a "data" folder in a locally cloned repo,
since it is included in the .gitignore file and won't automatically be added to the repo.
- Some gitignore documentation: https://help.github.com/articles/ignoring-files/
You can run the code using a python virtual environment (http://docs.python-guide.org/en/latest/dev/virtualenvs/).
virtualenv -p python3 venv
source venv/bin/activate
pip install -r requirements.txt
./run.sh
Here comes some general info about the data.
Ideas, from pre-processing to the end.
What we get out of it.