Batch vs real-time inference in NLP - Interactive Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to perform batch inference on a list of texts using a model.

NLP

predictions = model.predict([1])

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

Asentence

Btext

Ctexts

Dinput_text

Attempts:

3 left

💡 Hint

Common Mistakes

Passing a single string instead of a list causes errors or wrong predictions.

The model expects a list of texts for batch inference, so we pass texts.

2fill in blank

medium

Complete the code to perform real-time inference on a single input text.

NLP

result = model.predict([1])

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

Atexts

Btext

Cbatch

Dinputs

Attempts:

3 left

💡 Hint

Common Mistakes

Passing a list instead of a single string causes shape errors.

Real-time inference processes one input at a time, so we pass text.

3fill in blank

hard

Fix the error in the batch inference code by choosing the correct input format.

NLP

outputs = model.predict([1])

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

Atexts

B[text]

Ctext

Dtext.split()

Attempts:

3 left

💡 Hint

Common Mistakes

Passing a single string or a list with one string inside causes unexpected results.

Batch inference requires a list of texts, so texts is correct.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that maps each text to its prediction in batch mode.

NLP

results = {text: model.predict([1]) for text in [2]

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

A[text]

Btext

Ctexts

Dtext.split()

Attempts:

3 left

💡 Hint

Common Mistakes

Passing text directly without wrapping causes errors.

Iterating over a single string instead of the list of texts.

For each text in texts, we pass a list containing that single text [text] to model.predict because the model expects a list input.

5fill in blank

hard

Fill all three blanks to create a batch inference dictionary that filters texts longer than 5 words and maps them to predictions.

NLP

filtered_results = { [1]: model.predict([[2]]) for [3] in texts if len([1].split()) > 5 }

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

Atext

Dsentence

Attempts:

3 left

💡 Hint

Common Mistakes

Using different variable names inconsistently.

Not wrapping the input in a list for prediction.

We use text as the variable name consistently. The comprehension iterates over texts, filters texts longer than 5 words, and predicts on each wrapped in a list.