For Experts

Data Processing

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

Graph Compilation

• Process Control Statements
• Compiling Performance Optimization for Static Graph Network
• Calling the Custom Class
• Construct Constants In the Network
• Dependency Control

Model Training Optimization

• Sinking Mode
• Gradient Accumulation Algorithm
• Adaptive Gradient Summation Algorithm
• Dimension Reduction Training Algorithm
• Second-order Optimization

Custom Operator

• Custom Operators (Custom-based)
• MindSpore Hybrid Syntax Specification
• Advanced Usage of Custom Operators

Automatic Vectorization

• Automatic Vectorization (Vmap)

Model Inference

• Inference Model Overview
• Model Compression

Debugging and Tuning

• Function Debug
• Performance Tuning
• Precision Optimization↗

Distributed Parallel

• Distributed Parallel Training Mode
• Quick Start Distributed Parallel Training
• Distributed Set Communication Primitives
• Distributed Case
• Distributed Inference
• Saving and Loading Models in Hybrid Parallel Mode
• Distributed Fault Recovery
• Multi Dimensional
• Distributed Resilience Training and Inference
• Other Features

Environment Variables

• Environment Variables