Batch prediction vs real-time serving in MLOps - Interactive Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to perform batch prediction on a dataset using a trained model.

MLOps

predictions = model.[1](X_batch)

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

Atransform

Bfit

Cpredict

Dscore

Attempts:

3 left

💡 Hint

Common Mistakes

Using fit instead of predict.

Using transform which is for feature transformations.

The predict method is used to generate predictions from the model on a batch of input data.

2fill in blank

medium

Complete the code to serve a real-time prediction request using a deployed model.

MLOps

def serve_request(input_data):
    prediction = model.[1](input_data)
    return prediction

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

Apredict

Bevaluate

Ctrain

Dfit

Attempts:

3 left

💡 Hint

Common Mistakes

Using fit or train instead of predict.

Using evaluate which is for testing model performance.

For real-time serving, the model uses predict to generate a prediction for the incoming single input.

3fill in blank

hard

Fix the error in the batch prediction code by completing the blank.

MLOps

batch_predictions = model.predict([1])

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

Afit

By_batch

Cmodel

DX_batch

Attempts:

3 left

💡 Hint

Common Mistakes

Passing labels y_batch to predict.

Passing the model itself or calling fit instead.

The input to predict should be the batch features X_batch, not labels or other variables.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that maps input IDs to their batch predictions.

MLOps

id_to_prediction = {id: [1] for id, [2] in zip(ids, batch_data)}

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

Amodel.predict([batch_data])

Bmodel.predict({{BLANK_2}})

Cdata

Dfeatures

Attempts:

3 left

💡 Hint

Common Mistakes

Using model.predict([batch_data]) which passes a list of the whole batch.

Using a wrong variable name for the batch item.

The dictionary comprehension maps each id to the prediction from the model on the corresponding features.

5fill in blank

hard

Fill all three blanks to complete the real-time serving function that preprocesses input, predicts, and returns the result.

MLOps

def serve(input_raw):
    processed = [1](input_raw)
    prediction = model.[2](processed)
    return [3]

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

Apreprocess

Bpredict

Cprediction

Dtransform

Attempts:

3 left

💡 Hint

Common Mistakes

Returning the processed data instead of the prediction.

Using transform instead of predict for model output.

The function preprocesses the raw input, predicts using the model, and returns the prediction.