Data Science Model Deployments and Cloud Computing on GCP - Lab - Deploy Model Serving Code to App Engine

The video tutorial covers setting up a model serving environment using Python, Flask, and Google Cloud. It explains the main.py file, which loads a serialized model using pickle and handles POST requests for predictions. The app.yaml and requirements.txt files are discussed, followed by the deployment process using gcloud. Finally, the video demonstrates testing the model endpoint with JSON input, ensuring the application returns correct predictions.

Which library is used to serialize the model in the deployment setup?

• yaml

• xml

• json

• pickle

What type of request does the loadmodel function accept?

• DELETE

• PUT

• GET

• POST

Which Python version is specified in the app.yaml file for the runtime environment?

• Python 3.9

• Python 3.8

• Python 3.7

• Python 3.6

Which file contains the list of dependencies required for the deployment?

• requirements.txt

• config.json

• main.py

• app.yaml

What is the name of the service as specified in the app.yaml file?

• model_deployment

• prediction_service

• classification_model_prediction

• model_service

What command is used to deploy the application using Google Cloud?

• gcloud app deploy

• gcloud app start

• gcloud app run

• gcloud app launch

What is the expected status code if the POST request to the model endpoint is successful?

• 403

• 200

• 500

• 404