FastAPI for model serving in ML Python - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import FastAPI and create an app instance.

ML Python

from fastapi import [1]

app = [1]()

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

AResponse

BDepends

CRequest

DFastAPI

Attempts:

3 left

💡 Hint

Common Mistakes

Importing Request or Response instead of FastAPI

Forgetting to call FastAPI() as a function

We import FastAPI from the fastapi package and create an app instance by calling FastAPI().

2fill in blank

medium

Complete the code to define a POST endpoint '/predict' that accepts JSON input.

ML Python

@app.post('/predict')
async def predict(data: [1]):
    return {'prediction': 'dummy'}

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

Adict

Bstr

Cint

Dfloat

Attempts:

3 left

💡 Hint

Common Mistakes

Using str instead of dict for JSON input

Using numeric types like int or float incorrectly

The endpoint expects JSON data, which is parsed as a Python dictionary (dict).

3fill in blank

hard

Fix the error in the code to load a saved model using joblib.

ML Python

import joblib

model = joblib.[1]('model.joblib')

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

Asave

Bload

Cdump

Dopen

Attempts:

3 left

💡 Hint

Common Mistakes

Using save or dump instead of load

Trying to use open directly on the model file

To load a saved model with joblib, use joblib.load().

4fill in blank

hard

Fill both blanks to extract features from input JSON and make a prediction.

ML Python

features = [data[[1]] for [2] in ['feature1', 'feature2']]
prediction = model.predict([features])

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

A'feature1'

Bx

C'feature2'

Ddata

Attempts:

3 left

💡 Hint

Common Mistakes

Using a literal string instead of the loop variable in the first blank

Using a variable name that doesn't match the loop in the second blank

We extract each feature by its name from the input dictionary data using a loop variable x.

5fill in blank

hard

Fill all three blanks to return the prediction as JSON with a status code.

ML Python

from fastapi.responses import Response

@app.post('/predict')
async def predict(data: dict):
    features = [data[[1]] for [2] in ['f1', 'f2']]
    pred = model.predict([features])[0]
    return Response(content=str(pred), media_type=[3], status_code=200)

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

A'f1'

Bf

C'application/json'

D'text/plain'

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong feature key or loop variable

Using incorrect media type like 'application/json' instead of 'text/plain'

The loop variable f is used for feature extraction in both blanks, and the response media type for plain text is 'text/plain'.