Hello, learners!
Over the last few days, I’ve been diving into Docker, and I’m excited to share my journey with you. In this post, I’ll walk you through my experience from Day 2 and Day 3 of my Docker workshop and how I containerized a Python application.
Day 2: Getting Started with Docker
On Day 2, I got hands-on with Docker and learned the basics of containerization. Here's what I did:
Creating the Dockerfile:
I started by writing a Dockerfile to containerize a simple Python application. This involved using the official Python image, setting up a working directory, and installing the necessary dependencies (like scikit-learn, pandas, and joblib). The script is set to run when the container starts.
Building the Docker Image:
After writing the Dockerfile, I built my Docker image using the following command:
docker build -t my-python-app .
Running the Docker Container:
Once the image was built, I ran the container with the command:
docker run my-python-app
This allowed me to run the Python script in a contained, isolated environment, ensuring that the dependencies and environment were consistent.
Day 3: Implementing the KNN Algorithm
On Day 3, I continued my journey by working with a sales dataset to implement a K-Nearest Neighbors (KNN) algorithm.
Using a Sales Dataset:
I used a dataset to predict sales performance based on features like region, sales rep, and product category. This was a great exercise in both data science and Dockerization.
Creating the Python Script:
I wrote a Python script, sales.py, that applied the KNN algorithm to the sales dataset. The script used scikit-learn to train the model and make predictions.
Dockerizing the Sales Algorithm:
After developing the script, I Dockerized the sales.py application, just like I did with the previous Python script. This ensured that the application and all its dependencies were bundled into a container, making it easy to run anywhere.
Pushing the Docker Image to Docker Hub:
Finally, I pushed the custom Docker image for the sales algorithm to Docker Hub so I could easily share and access it from anywhere.
Commands I Used:
Building the Docker image:
docker build -t my-python-app .
Running the container:
docker run my-python-app
Pushing the image to Docker Hub:
docker push my-python-app
Takeaways
Docker is a game-changer when it comes to packaging applications and their dependencies in a portable and isolated environment.
The process of creating a Dockerfile, building images, and running containers is straightforward and very powerful for developers.
Docker Hub is an excellent platform for sharing and storing your Docker images.
Day 3: KNN Algorithm with Docker
