Docker for ML reproducibility in MLOps - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the Dockerfile line to specify the base image for a Python ML project.

MLOps

FROM [1]

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

Anode:14

Bubuntu:latest

Calpine:3.12

Dpython:3.9-slim

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing a base image without Python installed.

Using a Node.js image which is not suitable for Python ML projects.

The base image python:3.9-slim provides a lightweight Python environment ideal for ML projects.

2fill in blank

medium

Complete the command to build a Docker image named 'ml-model' from the current directory.

MLOps

docker build -t [1] .

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

Aml-model

Bmyapp

Cpython-app

Dmodel-image

Attempts:

3 left

💡 Hint

Common Mistakes

Using a generic or unrelated image name.

Omitting the dot at the end of the build command.

The -t flag tags the image with the name ml-model, which is used to identify it later.

3fill in blank

hard

Fix the error in the Dockerfile line to install dependencies from requirements.txt.

MLOps

RUN pip [1] -r requirements.txt

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

Adownload

Bupdate

Cinstall

Dupgrade

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'pip update' which is not a valid pip command.

Using 'pip upgrade' which does not install new packages.

The correct pip command to install packages is pip install -r requirements.txt.

4fill in blank

hard

Fill both blanks to run the container interactively and remove it after exit.

MLOps

docker run [1] [2] ml-model

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

A-it

B-rm

C--rm

D-d

Attempts:

3 left

💡 Hint

Common Mistakes

Using '-rm' instead of '--rm' which is invalid.

Using '-d' which runs the container detached, not interactively.

-it runs the container interactively with a terminal, and --rm removes the container after it stops.

5fill in blank

hard

Fill all three blanks to create a Docker volume, mount it to /data, and run the container using it.

MLOps

docker volume [1] ml_data

docker run -v ml_data:[2] [3] ml-model

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

Acreate

B/data

C--rm

Dremove

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'remove' instead of 'create' to make a volume.

Mounting to a wrong path inside the container.

Omitting the '--rm' flag causing leftover containers.

First, create the volume with docker volume create ml_data. Then mount it to /data inside the container and use --rm to remove the container after exit.