Performance metric tracking in MLOps - Mini Project: Build & Apply

Performance Metric Tracking

📖 Scenario: You are working on a machine learning project. You want to track the performance of your model by storing some key metrics like accuracy, precision, and recall.This helps you see how well your model is doing and compare different versions easily.

🎯 Goal: Build a simple Python script that stores performance metrics in a dictionary, sets a threshold for acceptable accuracy, filters metrics that meet the threshold, and finally prints the filtered metrics.

📋 What You'll Learn

Create a dictionary called metrics with exact keys and values for accuracy, precision, and recall.

Create a variable called accuracy_threshold with the value 0.8.

Use a dictionary comprehension to create a new dictionary called good_metrics that only includes metrics with values greater than or equal to accuracy_threshold.

Print the good_metrics dictionary.

💡 Why This Matters

🌍 Real World

Tracking performance metrics helps data scientists and engineers monitor how well machine learning models perform over time and after changes.

💼 Career

Knowing how to store, filter, and display performance metrics is essential for roles in MLOps, data engineering, and machine learning development.

1

Create the initial metrics dictionary

Create a dictionary called metrics with these exact entries: 'accuracy': 0.85, 'precision': 0.78, 'recall': 0.82.

MLOps

# Create the metrics dictionary with accuracy, precision, and recall
# Your code here

Need a hint?

Use curly braces {} to create a dictionary with the given keys and values.

2

Set the accuracy threshold

Create a variable called accuracy_threshold and set it to 0.8.

MLOps

metrics = {'accuracy': 0.85, 'precision': 0.78, 'recall': 0.82}
# Set the accuracy_threshold variable to 0.8
# Your code here

Need a hint?

Just assign the number 0.8 to the variable named accuracy_threshold.

3

Filter metrics using dictionary comprehension

Use a dictionary comprehension to create a new dictionary called good_metrics that includes only the entries from metrics where the value is greater than or equal to accuracy_threshold. Use metric and value as the loop variables.

MLOps

metrics = {'accuracy': 0.85, 'precision': 0.78, 'recall': 0.82}
accuracy_threshold = 0.8
# Create good_metrics dictionary with comprehension filtering values >= accuracy_threshold
# Your code here

Need a hint?

Use {metric: value for metric, value in metrics.items() if value >= accuracy_threshold} to filter the dictionary.

4

Print the filtered metrics

Write a print statement to display the good_metrics dictionary.

MLOps

metrics = {'accuracy': 0.85, 'precision': 0.78, 'recall': 0.82}
accuracy_threshold = 0.8
good_metrics = {metric: value for metric, value in metrics.items() if value >= accuracy_threshold}
# Print the good_metrics dictionary
# Your code here

Need a hint?

Use print(good_metrics) to show the filtered dictionary.