Scaling to Large Datasets with Criteo

Criteo provides the largest publicly available dataset for recommender systems, having a size of 1TB uncompressed click logs of 4 billion examples. We will teach to scale NVTabular:

• Using multiple GPUs and multiple nodes with NVTabular for ETL

• Training recommender system model with NVTabular dataloader for PyTorch

• Download and Convert
• ETL with NVTabular
• Training with TensorFlow
• Training with HugeCTR
• Training with FastAI
• Triton Inference with TensorFlow
• Triton Inference with HugeCTR