/KMML_challenge

Kernel Method for Machine Learning Kaggle challenge

Primary LanguageJupyter NotebookMIT LicenseMIT

KMML challenge

Kernel Method for Machine Learning Kaggle challenge

Starter pack

This code has been developed using Python 3.7.

The results are reproducible using (Mini)conda as follow (you may want to skip to step 4 if you have a running version of Python 3.7 that includes the requirements found in requirements.txt)

  1. To instanciate the conda environmment, run conda create --file conda-env.txt --name km
  2. Activate the virtual environement, using conda activate km (or source activate km)
  3. Then use pip to install the required Python packages in requirements.txt: pip -r requirements.txt
  4. Launch the main file: python start.py and grab a cup of coffee as it takes around 1 hour to complete :)
  5. You can find the final results in Yte.csv inside the results folder.