For Experts

Data Processing

• Auto Augmentation
• Lightweight Data Processing
• Single-Node Tensor Cache
• Optimizing the Data Processing

Operator Execution

• Operators Classification
• Operation Overloading
• Custom Operators (Ascend)
• Custom Operators (Custom based)

Model Inference

• Inference Model Overview
• Inference on a GPU
• Inference on the Ascend 910 AI processor
• Inference Using the MindIR Model on Ascend 310 AI Processors
• Inference on the Ascend 310 AI Processor

Debugging and Tuning

• Reading IR
• Using Dump in the Graph Mode
• Custom Debugging Information
• Incremental Operator Build
• AutoTune
• Dataset AutoTune for Dataset Pipeline

Distributed Parallel

• Distributed Parallel Training Mode
• Parallel Distributed Training Example (Ascend)
• Distributed Parallel Training Example (GPU)
• Distributed Inference
• Saving and Loading Models in Hybrid Parallel Mode

Advanced Features

• Enabling Mixed Precision
• Gradient Accumulation Algorithm