# Importing necessary modules from the surprise library
from surprise import KNNBasic
from surprise import Dataset
from surprise import Reader
from surprise.model_selection import cross_validate

# Pandas is used for data manipulation
import pandas as pd

# Load the ratings dataset
df = pd.read_csv("ratings.csv")

# Create a Reader object with the rating scale from 1 to 5
# The Reader class is used to parse a file containing ratings
reader = Reader(rating_scale=(1, 5))

# Load the data from the dataframe into the surprise Dataset format
# The load_from_df method loads a dataset from a pandas dataframe
# The dataframe must have three columns: user ids, item ids and ratings, in that order
data = Dataset.load_from_df(df[['userID', 'itemID', 'rating']], reader)

# Define the algorithm to be used, in this case, KNNBasic
algo = KNNBasic()

# Use cross-validation to assess the performance of the algorithm
# The cross_validate function performs a cross-validation procedure on a dataset with a given algorithm
# RMSE (Root Mean Squared Error) and MAE (Mean Absolute Error) are computed as accuracy metrics
# cv=5 implies a 5-fold cross-validation
# verbose=True will print messages (like progress and computed measures) during execution
cross_validate(algo, data, measures=['RMSE', 'MAE'], cv=5, verbose=True)