Binary model performance
You've already built several ALS models, so we won't do that again. An implicit ALS model has already been fitted to the binary ratings of the MovieLens dataset. Let's look at the binary_test_predictions
from this model to see what we can learn.
The ROEM()
function has been defined for you. Feel free to run help(ROEM)
in the console if you want more details on how to execute it!
This exercise is part of the course
Building Recommendation Engines with PySpark
Exercise instructions
- Import the
col
function from thepyspark.sql.functions
class. - Take a look at
binary_test_predictions
using the.show()
method to get an understanding of what the data looks like. - Call
ROEM()
on thebinary_test_predictions
to evaluate the model's performance. Do you think it performed well? - Use
.filter()
to look only at user 42's predictions (col("userId") == 42
). Do you notice anything?
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Import the col function
from ____.sql.____ import ____
# Look at the test predictions
____.____()
# Evaluate ROEM on test predictions
____(____)
# Look at user 42's test predictions
binary_test_predictions.____(col("____") == ____).show()