Basic Model Tracking¶

In this tutorial we'll show how to train a basic sci-kit learn model and log it into MLflow. First we need a vasic dataset:

In [17]:

import numpy as np


X = np.arange(100).reshape(100, 1)
y = X ** 2

import numpy as np X = np.arange(100).reshape(100, 1) y = X ** 2

Then select a simple linear regression model

In [ ]:

from sklearn.linear_model import LinearRegression


model = LinearRegression()

from sklearn.linear_model import LinearRegression model = LinearRegression()

Then we connect to the MLFlow server.

In [5]:

import mlflow


mlflow.set_tracking_uri("http://localhost:5000")

import mlflow mlflow.set_tracking_uri("http://localhost:5000")

Then we set the experiment.

In [6]:

experiment = mlflow.set_experiment("test")

experiment = mlflow.set_experiment("test")

Then we'll use autolog to simplify the process of logging the model and the metrics into MLflow,

In [7]:

mlflow.sklearn.autolog()

mlflow.sklearn.autolog()

Finally train the model and automatically log the metrics and the model.

In [9]:

experiment_id = experiment.experiment_id
with mlflow.start_run(experiment_id=experiment_id):
    model.fit(X, y)

experiment_id = experiment.experiment_id with mlflow.start_run(experiment_id=experiment_id): model.fit(X, y)

2024/10/12 17:59:03 INFO mlflow.tracking._tracking_service.client: 🏃 View run bittersweet-koi-963 at: http://localhost:5000/#/experiments/406653565916109575/runs/ef588dbfe5bd487abc3ca25eb5151e6f.
2024/10/12 17:59:03 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://localhost:5000/#/experiments/406653565916109575.