Weights and Biases overview in MLOps - Mini Project: Build & Apply

Weights and Biases Overview

📖 Scenario: You are working on a machine learning project and want to track your experiments easily. You decide to use Weights and Biases (W&B), a tool that helps you log your training runs, visualize results, and compare different models.

🎯 Goal: Learn how to set up a simple W&B project, configure it, log some metrics during training, and display the logged results.

📋 What You'll Learn

Create a Python dictionary with experiment parameters

Add a configuration dictionary for W&B setup

Log training metrics using a loop

Print the final logged metrics summary

💡 Why This Matters

🌍 Real World

Weights and Biases is widely used by data scientists and ML engineers to track experiments, compare models, and share results easily.

💼 Career

Knowing how to set up and use W&B is a valuable skill for roles in machine learning operations (MLOps) and data science teams.

1

Create experiment parameters dictionary

Create a dictionary called experiment_params with these exact entries: 'learning_rate': 0.01, 'batch_size': 32, 'epochs': 5

MLOps

# Your code here

Need a hint?

Use curly braces to create a dictionary with the exact keys and values.

2

Add W&B configuration dictionary

Create a dictionary called wandb_config with these exact entries: 'project': 'mlops-demo', 'entity': 'your-team'

MLOps

experiment_params = {'learning_rate': 0.01, 'batch_size': 32, 'epochs': 5}
# Your code here

Need a hint?

Create another dictionary with the exact keys and values for W&B setup.

3

Log training metrics with a loop

Use a for loop with variable epoch in range(1, experiment_params['epochs'] + 1) to simulate training. Inside the loop, create a dictionary metrics with 'epoch' set to epoch and 'accuracy' set to epoch * 0.1.

MLOps

experiment_params = {'learning_rate': 0.01, 'batch_size': 32, 'epochs': 5}
wandb_config = {'project': 'mlops-demo', 'entity': 'your-team'}
# Your code here

Need a hint?

Use a for loop to simulate epochs and create a metrics dictionary inside it.

4

Print the final metrics dictionary

After the loop, write print(metrics) to display the last epoch's metrics dictionary.

MLOps

experiment_params = {'learning_rate': 0.01, 'batch_size': 32, 'epochs': 5}
wandb_config = {'project': 'mlops-demo', 'entity': 'your-team'}
for epoch in range(1, experiment_params['epochs'] + 1):
    metrics = {'epoch': epoch, 'accuracy': epoch * 0.1}
# Your code here

Need a hint?

Print the metrics dictionary after the loop to see the final epoch's data.