Usage

In order to use spare_scores, you have to use our CLI. The CLI can perform training and testing. For training we use 2 models, a SVM and a MLP. To perform training, you can just do:

$ spare_score --action train \
    --input spare_scores/data/example_data.csv \
    --predictors H_MUSE_Volume_11 H_MUSE_Volume_23 H_MUSE_Volume_30 \
    --ignore_vars Sex \
    --to_predict Age \
    --kernel linear \
    --verbose 2 \
    --output my_model.pkl.gz

With the --action parameter, you specify if you want to perform training or testing.
With the --input parameter, you specify the directory of the input data(has to be .csv).
The --predictors parameter is a list that represents the columns that will be used by the models for training.
With the --ignore_vars parameter, you specify(if needed) any columns than you want the models to ignore.
--to_predict represents the target column.
--kernel represents the kernel of regression/classification. Currently only linear is supported as an option for regression.
--verbose you can pass a value != 0 to enable verbosity on training/testing.
With the --output parameter, you specify the directory of the output. This is where the trained model will be saved.

To perform testing, you can just do:

$ spare_score -a test \
    -i spare_scores/data/example_data.csv  \
    --model my_model.pkl.gz \
    -o test_spare_data.csv \
    -v 0 \
    --logs test_logs.txt

The only new parameter here is --logs that represents the filename of the logger.