For Experts

Data Processing

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

Graph Compilation

• Process Control Statements
• Compiling Performance Optimization for Static Graph Network
• Customizing Reverse Propagation Function of Cell
• Calling the Custom Class
• Construct Constants In the Network
• Dependency Control

Model Training Optimization

• Enabling Mixed Precision
• Gradient Accumulation Algorithm
• Adaptive Gradient Summation Algorithm
• Dimension Reduction Training Algorithm

Custom Operator

• 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

• Function Debug
• Performance Tuning
• Precision Optimization↗

Distributed Parallel

• Distributed Parallel Training Mode
• Distributed Case
• Distributed Inference
• Saving and Loading Models in Hybrid Parallel Mode
• Multi Dimensional
• Other Features

Environment Variables

• Environment Variables