MLflow experiments
MLflow experiments are used as a way to organize data from training runs in a way that can be easily searched and queried for our analysis later.
In this exercise, you will use the MLflow module to create a new experiment called Unicorn Model
for your new ML project. You will add useful information to the experiment by setting tags for the version. Finally, you will set the Unicorn Model
experiment as your current experiment so when you begin tracking your data will be tracked within this particular experiment.
This exercise is part of the course
Introduction to MLflow
Exercise instructions
- Import the MLflow module.
- Create a new experiment called
"Unicorn Model"
. - On the
Unicorn Model
, set the set the tags as"version"
and"1.0"
. - Set the experiment
"Unicorn Model"
as the current experiment for tracking.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Import MLflow
import ____
# Create new experiment
mlflow.____("____ ____")
# Tag new experiment
mlflow.____("____", "____")
# Set the experiment
mlflow.____("____ ____")