name: houses-ds-project
conda_env: environment.yml
entry_points:
  fetch_dataset:
    parameters:
      output_dir: {
        type: string,
        default: data/raw
      }
    command: >
          kaggle competitions download -c house-prices-advanced-regression-techniques -p {output_dir} ;\
          unzip -o {output_dir}/*.zip -d {output_dir}/

  main:
    parameters:
      input_filepath: {
        type: string,
        default: data/raw/train.csv
      }
      output_filepath: {
        type: string,
        default: "data/interim/train.csv"
      }
      redundant_columns_threhold: {
        type: float, default: 0.9
      }
      nan_columns_threshold: {
        type: float, default: 0.9
      }
      rare_categories_threshold: {
        type: float, default: 0.1
      }
      verbose: {
        type: boolean,
        default: False
      }

    command: >
      python -m houses_pipeline.preprocess {input_filepath} {output_filepath} \
        --redundant_columns_threhold={redundant_columns_threhold} \
        --nan_columns_threshold={nan_columns_threshold} \
        --rare_categories_threshold={rare_categories_threshold} \
        --verbose={verbose}
