Word2Vec (CBOW and Skip-gram) in NLP - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the Word2Vec model from gensim.

NLP

from gensim.models import [1]

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

AWord2Vec

BFastText

CDoc2Vec

DLdaModel

Attempts:

3 left

💡 Hint

Common Mistakes

Importing unrelated models like FastText or Doc2Vec instead of Word2Vec.

The Word2Vec class is imported from gensim.models to create word embedding models.

2fill in blank

medium

Complete the code to initialize a CBOW Word2Vec model with vector size 100.

NLP

model = Word2Vec(sentences, vector_size=[1], window=5, sg=0, min_count=1)

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

A50

B200

C100

D300

Attempts:

3 left

💡 Hint

Common Mistakes

Using vector_size values that are too small or too large without reason.

The vector_size parameter sets the size of the word vectors. 100 is a common choice.

3fill in blank

hard

Fix the error in the code to train a Skip-gram Word2Vec model.

NLP

model = Word2Vec(sentences, vector_size=100, window=5, sg=[1], min_count=1)

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

A0

B1

C2

D-1

Attempts:

3 left

💡 Hint

Common Mistakes

Using sg=0 which trains CBOW instead of Skip-gram.

Using invalid values like 2 or -1 for sg.

Setting sg=1 configures the model to use the Skip-gram architecture.

4fill in blank

hard

Fill both blanks to create a dictionary of word vectors for words with frequency above 2.

NLP

word_vectors = {word: model.wv[[1]] for word in model.wv.index_to_key if model.wv.get_vecattr(word, '[2]') > 2}

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

Aword

Bcount

Cfrequency

Dindex

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'frequency' instead of 'count' for the attribute name.

Using 'index' instead of the actual word to get vectors.

We use 'word' to get the vector and 'count' to check the frequency attribute.

5fill in blank

hard

Fill all three blanks to find the top 3 most similar words to 'king'.

NLP

similar_words = model.wv.most_similar(positive=[[1]], topn=[2])
result = [word for word, [3] in similar_words]

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

A'king'

B3

Csimilarity

D'queen'

Attempts:

3 left

💡 Hint

Common Mistakes

Using the wrong variable name instead of 'similarity' in the unpacking.

Passing a word not in quotes to the positive list.

We search for words similar to 'king', get top 3, and extract words ignoring similarity scores.