reproduce_critiquing.py

from experiment.critiquing import critiquing
from utils.argcheck import check_int_positive

import ast
import argparse
import pandas as pd


def main(args):

    num_users = pd.read_csv(args.data_dir + args.user_col + '.csv')[args.user_col].nunique()
    num_items = pd.read_csv(args.data_dir + args.item_col + '.csv')[args.item_col].nunique()

    df_train = pd.read_csv(args.data_dir + args.train_set)
    df_train = df_train[df_train[args.rating_col] == 1]
    df_train[args.keyphrase_vector_col] = df_train[args.keyphrase_vector_col].apply(ast.literal_eval)

    keyphrase_names = pd.read_csv(args.data_dir + args.keyphrase_set)[args.keyphrase_col].values

    params = dict()
    params['model_saved_path'] = args.model_saved_path

    critiquing(num_users,
               num_items,
               args.user_col,
               args.item_col,
               args.rating_col,
               args.keyphrase_vector_col,
               df_train,
               keyphrase_names,
               params,
               args.num_users_sampled,
               load_path=args.load_path,
               save_path=args.save_path)


if __name__ == "__main__":
    # Commandline arguments
    parser = argparse.ArgumentParser(description="Reproduce Critiquing Performance")

    parser.add_argument('--data_dir', dest='data_dir', default="data/CDsVinyl/")
    parser.add_argument('--item_col', dest='item_col', default="ItemIndex")
    parser.add_argument('--keyphrase', dest='keyphrase_set', default="KeyPhrases.csv")
    parser.add_argument('--keyphrase_col', dest='keyphrase_col', default="Phrases")
    parser.add_argument('--keyphrase_vector_col', dest='keyphrase_vector_col', default="keyVector")
    parser.add_argument('--load_path', dest='load_path', default="explanation/CDsVinyl/hyper_parameters.csv")
    parser.add_argument('--model_saved_path', dest='model_saved_path', default="CDsVinyl")
    parser.add_argument('--num_users_sampled', dest='num_users_sampled', type=check_int_positive, default=500)
    parser.add_argument('--rating_col', dest='rating_col', default="Binary")
    parser.add_argument('--save_path', dest='save_path', default="CD_Critiquing")
    parser.add_argument('--train', dest='train_set', default="Train.csv")
    parser.add_argument('--user_col', dest='user_col', default="UserIndex")

    args = parser.parse_args()

    main(args)