Tune experiment

With the Tune experiment function, the platform suggests a couple of changes that might improve your model’s performance. The only way to find out what works best is to test.

You can create the following adjusted experiments:

How to Tune experiment

  1. Click the Tune experiment button in the Evaluation view after one epoch while the model is running or when the model is done training. Tune experiment button

  2. Select the experiments you want to create and click Create and run.
    The new experiment will keep the tags of the first experiment. Create and run button

Other ways to continue experimenting

  • Duplicate the experiment and make your own changes to the model.
    When duplicating, you copy the complete project without weights.

  • Iterate on an experiment to resume training with different settings.
    When iterating, you keep the weights. That is, you keep what the model has learned previously.

    • Continue training Use when an experiment showed really good progress up to a certain checkpoint but then it wasn’t so good anymore.

    • Reuse part of model lets you reuse a part of your model with weights from a specific checkpoint in a new model. For example, when you want to train a model with additional datasets.

Tips on how to improve an experiment

Suppose you want some tips on how to improve your results. In that case, you can probably find some inspiration in our articles on how to improve experiments—both for beginners and for intermediate users.

Was this page helpful?