MLflow setup and basics in MLOps - Mini Project: Build & Apply

MLflow Setup and Basics

📖 Scenario: You are a data scientist starting a new machine learning project. You want to keep track of your experiments, parameters, and results easily. MLflow is a tool that helps you do this by logging your work in a simple way.

🎯 Goal: Set up MLflow in a Python script, log parameters and metrics for a simple experiment, and display the logged information.

📋 What You'll Learn

Create a Python dictionary called params with exact keys and values

Create a variable called metric_value with a specific float value

Use MLflow to start a run and log the parameters and metric

Print the logged parameters and metric values

💡 Why This Matters

🌍 Real World

MLflow helps data scientists keep track of their machine learning experiments easily, so they can compare results and reproduce work.

💼 Career

Knowing how to use MLflow is valuable for roles in machine learning engineering and data science, as it improves collaboration and experiment management.

1

Create experiment parameters

Create a Python dictionary called params with these exact entries: 'learning_rate': 0.01, 'num_trees': 100, and 'max_depth': 5.

MLOps

# Your code here

Need a hint?

Use curly braces to create a dictionary and separate key-value pairs with commas.

2

Set a metric value

Create a variable called metric_value and set it to the float 0.85.

MLOps

params = {'learning_rate': 0.01, 'num_trees': 100, 'max_depth': 5}
# Your code here

Need a hint?

Assign the float value directly to the variable.

3

Log parameters and metric with MLflow

Import mlflow. Use mlflow.start_run() as a context manager. Inside it, log the parameters from params using mlflow.log_params(params) and log the metric metric_value with the name 'accuracy' using mlflow.log_metric('accuracy', metric_value).

MLOps

params = {'learning_rate': 0.01, 'num_trees': 100, 'max_depth': 5}
metric_value = 0.85
# Your code here

Need a hint?

Use with mlflow.start_run(): to start logging, then call the logging functions inside.

4

Print logged parameters and metric

Print the params dictionary and the string 'Accuracy:' followed by the metric_value variable on the next line.

MLOps

import mlflow

params = {'learning_rate': 0.01, 'num_trees': 100, 'max_depth': 5}
metric_value = 0.85

with mlflow.start_run():
    mlflow.log_params(params)
    mlflow.log_metric('accuracy', metric_value)
# Your code here

Need a hint?

Use two print statements: one for the dictionary and one for the metric with label.