Skip to main content
Version: v1.4.0

Tutorials

Learn how to train state-of-the-art deep neural networks on a large set of diverse problem types

Learning path​

note

To learn how H2O Hydrogen Torch can help build and deploy a model with H2O Label Genie and H2O MLOps, refer to the following blog: In H2O AI Cloud, build, deploy, and score a state-of-the-art image classification model, starting with unlabeled data.

Experiments​

  • Tutorial 1A: Default hyperparameter values

    This tutorial builds an Image regression model capable of predicting the sum of coins in images. While building and training the model, this tutorial focuses on how you can generate fairly accurate machine learning models with default hyperparameter values derived from model training best practices used by top Kaggle grandmasters.

    Also, this tutorial highlights available charts to understand the impact of selected hyperparameter values on the model's training.

  • Tutorial 2A: Model tuning with grid search

    This tutorial aims to demonstrate the benefits of using grid search in H2O Hydrogen Torch to improve a pre-existing model. Specifically, we will focus on building and refining an image metric learning model that can effectively determine the similarity or dissimilarity between bicycle images.

    By following this tutorial, you will gain a deeper understanding of how grid search can:

    1. Fine-tune and enhance models created in H2O Hydrogen Torch.
    2. Streamline and eliminate the repetitive process of creating multiple models with varying hyperparameter values.
    3. Elevate your built models in H2O Hydrogen Torch from being merely good to state-of-the-art.

Predictions​

  • Tutorial 1B: Model deployment in the H2O Hydrogen Torch UI

    This tutorial explores one of the options available to deploy a built model. In particular, this tutorial explores how to deploy a built model in the H2O Hydrogen Torch UI. This tutorial also explores downloading the generated predictions and how you can view the downloaded predictions using Python. At a high level, we first create a model that you will later deploy to score on new data.

  • Tutorial 2B: Model deployment with a model's H2O MLOps pipeline

    This tutorial explores one of the options available to deploy a built model. In particular, this tutorial builds an image regression model to explore how it can be deployed to H2O MLOps using the model's H2O MLOps pipeline.

  • Tutorial 3B: Model deployment with a model's Python scoring pipeline

    This tutorial explores one of the options available to deploy a built model. In particular, this tutorial builds an image regression model to explore how you can deploy a model in any external Python environment using the model's standalone Python scoring pipeline.

  • Tutorial 4B: Model deployment to H2O MLOps through the H2O Hydrogen Torch UI

    This tutorial explores one of the options available to deploy a built model. In particular, this tutorial builds an image regression model to explore how you can deploy a model to H2O MLOps through the H2O Hydrogen Torch UI.

Feedback