Embedding models for semantic search in Agentic AI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create an embedding vector from text using a model.

Agentic AI

embedding = model.[1](text)

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

Atrain

Bdecode

Cencode

Dpredict

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'decode' instead of 'encode' which is for reversing embeddings.

Using 'train' which is for model training, not embedding generation.

The encode method converts text into an embedding vector for semantic search.

2fill in blank

medium

Complete the code to compute cosine similarity between two embedding vectors.

Agentic AI

similarity = cosine_similarity(vec1, [1])

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

Avec2

Bmodel

Cembedding

Dtext

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the model or raw text instead of the second vector.

Using the embedding variable without specifying which vector.

Cosine similarity compares two vectors, so the second vector vec2 is needed.

3fill in blank

hard

Fix the error in the code to normalize an embedding vector.

Agentic AI

normalized_vec = vec / np.[1](vec)

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

Amax

Bmean

Csum

Dlinalg.norm

Attempts:

3 left

💡 Hint

Common Mistakes

Using sum or mean which do not give vector length.

Using max which only finds the largest element.

Normalization divides by the vector's length, computed by np.linalg.norm.

4fill in blank

hard

Fill both blanks to create a dictionary of words and their embedding lengths greater than 5.

Agentic AI

lengths = {word: [1] for word in words if [2] > 5}

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

Alen(word)

Bword

Dembedding[word]

Attempts:

3 left

💡 Hint

Common Mistakes

Using the word itself as value instead of embedding length.

Confusing word length with embedding vector length.

The dictionary maps each word to the length of its embedding vector, filtering words with length greater than 5.

5fill in blank

hard

Fill all three blanks to filter embeddings with similarity above 0.8 and create a result dictionary.

Agentic AI

result = [1]: [2] for [3] in embeddings if similarity(embeddings[query], embeddings[[1]]) > 0.8}

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

Aword

Bembeddings[word]

Ditem

Attempts:

3 left

💡 Hint

Common Mistakes

Using incorrect variable names causing runtime errors.

Mixing keys and values in the dictionary comprehension.

The dictionary comprehension uses word as key and its embedding as value, iterating over words in embeddings and filtering by similarity.