AI experimentation¶
The following sections provide details for creating Workbench machine learning and time series experiments:
| Topic | Description |
|---|---|
| Create and evaluate predictive experiments | Create non-time series (predictive) experiments, both supervised or unsupervised. Fine-tune experiment setup, making row-by-row predictions based on your data. |
| Create and evaluate time-aware experiments | Create supervised or unsupervised experiments and fine-tune experiment setup, using time-relevant data to make row-by-row predictions, time series forecasts, or current value predictions "nowcasts". |
| Manage experiments | Access the data insights and the model Leaderboard to evaluate and compare models and experiments. |
| Evaluate with model insights | View model insights to interpret, explain, and validate what drives a model predictions. |
| Make predictions | After you create an experiment and train models, you can upload scoring data, make predictions, and download the results. |
| Experiment reference | Provide reference content that supports working with predictive and time-aware experiments. |