scikit-learn with Joblib
The sklearn_joblib_example.py demonstrates integrating Trains into code which uses scikit-learn, joblib to store a model and model snapshot, and matplotlib to create a scatter diagram. When the script runs, it creates an experiment named scikit-learn joblib example which is associated with the examples project.
Trains automatically logs the scatter plot, which appears in the RESULTS tab, PLOTS sub-tab.
Trains stores the model and the experiment separately, allowing you to use models in different experiments not just the experiment creating the model. The Trains Web (UI), therefore, shows the input model in the experiment details separate from the model details (above).
The input model appears in the ARTIFACTS tab, Input Model area, where the model name is a hyperlink to the model details.
Trains automatically stores the model and model snapshot. They appear in the model details GENERAL tab, where you can see the following:
- Input model location (URL)
- Model snapshots / checkpoint model locations (URLs).
- Experiment creating the model.
- Other general information about the model.
As in all code which contains initializes a Task, Trains automatically logs the console output.