def main():
paths = "/home/nikoscf/PycharmProjects/BookRecommendation/configurations/paths.yml"
load_begin = DataLoader()
load_begin.read_paths(paths)
# Uncomment to Execute this one time to get the zip if is .zip, unzip it in absolute dir you set in paths.yaml
# Then it checks for .csv and remove the redundant zip folder
# load_begin.check_zip_and_csv()
books = load_begin.read_data("BX-Books.csv")
users = load_begin.read_data("BX-Users.csv")
ratings = load_begin.read_data("BX-Book-Ratings.csv")
to_drop_columns = ['Image-URL-S', 'Image-URL-M', 'Image-URL-L']
numeric_col_to_nan = ["Year-Of-Publication"]
data_books = DataClean(books)
clean_books = data_books.execute_pipeline_cleaning(to_drop_columns,
numeric_col_to_nan)
to_drop_columns = []
numeric_col_to_nan = ["User-ID", "Age"]
data_users = DataClean(users)
clean_users = data_users.execute_pipeline_cleaning(to_drop_columns,
numeric_col_to_nan)
to_drop_columns = []
numeric_col_to_nan = ["User-ID", "ISBN", "Book-Rating"]
data_ratings = DataClean(ratings)
clean_ratings = data_ratings.execute_pipeline_cleaning(
to_drop_columns, numeric_col_to_nan)
data_analysis = DataAnalysis()
ratings_pivoted = data_analysis.execute_pipeline_data_analysis(
clean_ratings, clean_users, clean_books)
return ratings_pivoted
