Multi-query retrieval in Prompt Engineering / GenAI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a list of queries for multi-query retrieval.

Prompt Engineering / GenAI

queries = ["What is AI?", [1], "Explain machine learning."]

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

ANone

B42

C"How does retrieval work?"

DTrue

Attempts:

3 left

💡 Hint

Common Mistakes

Using non-string values as queries.

Leaving the blank empty.

The list of queries must contain strings representing questions. Option C is a valid query string.

2fill in blank

medium

Complete the code to initialize the retrieval model for multi-query retrieval.

Prompt Engineering / GenAI

model = RetrievalModel([1]=True)

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

Abatch_size

Bmulti_query

Clearning_rate

Dnum_epochs

Attempts:

3 left

💡 Hint

Common Mistakes

Using unrelated parameters like batch_size or learning_rate.

Misspelling the parameter name.

The parameter 'multi_query=True' enables multi-query retrieval in the model initialization.

3fill in blank

hard

Fix the error in the code to correctly perform multi-query retrieval.

Prompt Engineering / GenAI

results = model.retrieve(queries=[1], top_k=5)

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

Aqueries

Bquery

Cquery_list

Dquerys

Attempts:

3 left

💡 Hint

Common Mistakes

Using singular 'query' instead of 'queries'.

Misspelling the parameter name.

The method expects the argument named 'queries' to receive the list of queries.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that maps each query to its top result.

Prompt Engineering / GenAI

top_results = {q: results[[1]][[2]] for q in queries}

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

Aq

B0

C1

Dresults

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong keys or indices to access results.

Confusing the query variable with the results variable.

For each query q, results[q][0] gives the top result. So BLANK_1 is 'q' and BLANK_2 is '0'.

5fill in blank

hard

Fill all three blanks to filter queries and create a dictionary of their top results only if the score is above 0.8.

Prompt Engineering / GenAI

filtered_results = {q: results[q][[1]] for q in queries if results[q][[2]] [3] 0.8}

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

A0

B1

C>

D<

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing up indices for result and score.

Using wrong comparison operators.

Index 0 accesses the top result, index 1 accesses the score. We filter where score > 0.8.