Merlin PyTorch Jupyter Notebooks

These example notebooks demonstrate how to use NVTabular with HugeCTR. Each example provides details about the end-to-end workflow, which includes Download data, ETL, and Training.

About Quick Deploy

The quick deploy feature automatically sets up the Vertex AI instance with an optimal configuration, preloads the dependencies, runs the software from NGC without any need to set up the infrastructure.

Each example notebook is structured as follows:

1. 01-Download-Convert.ipynb: Demonstrates how to download the dataset and convert it into the correct format so that it can be consumed.
2. 02-ETL-with-NVTabular.ipynb: Demonstrates how to execute the preprocessing and feature engineering pipeline (ETL) with NVTabular on the GPU.
3. 03-Training-with-PyTorch.ipynb: Demonstrates how to train a model with PyTorch based on the ETL output.
4. T4R-01-ETL-with-NVTabular.ipynb: Demonstrates how to execute the preprocessing and feature engineering pipeline (ETL) with NVTabular towards using a Transformer-based model.
5. T4R-02-session-based-XLNet-with-PyT.ipynb: Demonstrates use of Transformers4Rec library for session-based recommendation with synthetic data.

Deploying the notebooks

To deploy these notebooks with the optimal configuration to the cloud please click the "deploy" button on the top right of NGC.

Reading the notebook without leaving NGC

If you want to read through the notebook example without leaving our website, follow these steps:

1. Navigate to the File Browser tab of the asset in NGC
2. Select the version you'd like to see
3. Next to the .ipynb file select "View Jupyter"
4. There you have it! You can read a notebook for documentation and copy code samples without ever leaving NGC.
5. All the instructions you need to get started are in the resource - so head over and see how to get up and running.