Build Your First Image Regression Model in Hydrogen Torch
- Predictive AI
- H2O Hydrogen Torch
This lesson walks through creating a first image regression experiment in H2O Hydrogen Torch using default hyperparameter values. It covers selecting the coins_image_regression dataset, naming the experiment, and running it with settings the platform pre-selects based on established best practices. It explains how to read the results using Mean Absolute Error as the default scorer, where a value near zero means the model's predictions are close to the true sums, and points to the View experiments card for tracking running and completed runs.
Key takeaways
- How H2O Hydrogen Torch pre-selects default settings for an image regression experiment based on the dataset type
- Why a Mean Absolute Error close to zero indicates the model's predictions are close to the true values
- The View experiments card where running and completed Hydrogen Torch runs are tracked
Start owning your AI