mf_pyspark_als.py

mf_pyspark_als.py

from pyspark.ml.recommendations import ALS

# Context , there is dfs_actions is spark dataframe that looks like the pandas dataframe for surprise example https://gist.github.com/jeanmidevacc/a00c9cf8fe9379cd8a818b1d842dbaa1

# Setup the model parameters
als = ALS(
  seed=12,
  userCol="user_id",
  itemCol="item_id",
  ratingCol="rating",
  implicitPrefs=False,# to run the model for implicit rating and leverage
  nonnegative=False,
  maxIter=10,
  rank=10)

model = als.fit(dfs_actions)#Build the factors matrices

dfs_predictions = model.transform(dfs_actions)#Compute the prediction between the pair user_id and item_id

user_factors, item_factors = model.userFactors, model.itemFactors #To extract the factors from the model (careful they are only defined by id and not your column atribute in this user_id, item_id