Why recommendations drive engagement in ML Python - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the library used for building recommendation models.

ML Python

import [1]

Drag options to blanks, or click blank then click option'

Anumpy

Bsklearn

Ctensorflow

Dpandas

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing sklearn which is mainly for classical ML, not deep learning.

Choosing numpy or pandas which are for data manipulation, not modeling.

TensorFlow is commonly used to build recommendation models because it supports deep learning architectures.

2fill in blank

medium

Complete the code to create a simple user-item interaction matrix using pandas.

ML Python

import pandas as pd
interactions = pd.DataFrame({'user_id': [1, 2, 1], 'item_id': [101, 102, 103], 'rating': [5, 3, 4]})
matrix = interactions.pivot(index='user_id', columns=[1], values='rating')

Drag options to blanks, or click blank then click option'

A'item_id'

B'rating'

C'user_id'

D'interaction'

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'user_id' as columns which would not create the correct matrix.

Using 'rating' which is the values, not columns.

The pivot uses 'item_id' as columns to create the user-item matrix.

3fill in blank

hard

Fix the error in the code that trains a simple recommendation model using TensorFlow.

ML Python

import tensorflow as tf
model = tf.keras.Sequential([
    tf.keras.layers.Dense(64, activation='relu'),
    tf.keras.layers.Dense(1)
])
model.compile(optimizer='adam', loss=[1])
model.fit(x_train, y_train, epochs=5)

Drag options to blanks, or click blank then click option'

A'binary_crossentropy'

B'mse'

C'categorical_crossentropy'

D'accuracy'

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'accuracy' which is for classification, not regression.

Using 'categorical_crossentropy' which is for multi-class classification.

Mean squared error ('mse') is appropriate for regression tasks like rating prediction.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that maps items to their average ratings above 3.

ML Python

average_ratings = {item: [1] for item, ratings in ratings_dict.items() if [2] > 3}

Drag options to blanks, or click blank then click option'

Asum(ratings)/len(ratings)

Bmax(ratings)

Csum(ratings)

Dlen(ratings)

Attempts:

3 left

💡 Hint

Common Mistakes

Using max or sum alone which do not give average.

Using length alone which is count, not average.

We calculate the average rating with sum divided by length and filter items with average rating above 3.

5fill in blank

hard

Fill all three blanks to create a dictionary of recommended items with predicted scores above 4.

ML Python

recommendations = [1] for item, score in predicted_scores.items() if score [2] 4}
sorted_recs = sorted(recommendations.items(), key=[3], reverse=True)

Drag options to blanks, or click blank then click option'

A{item: score

B>

Clambda x: x[1]

D[item, score

Attempts:

3 left

💡 Hint

Common Mistakes

Using list brackets instead of dictionary braces.

Using '<' instead of '>' causing wrong filtering.

Sorting by key instead of value.

The dictionary comprehension collects items with scores greater than 4, then sorts by score descending.