Learning Roadmap

Jump to bottom

Sherlock edited this page Mar 12, 2021 · 1 revision

Basic

InferenceSession/ORTModule
Graph/Node/NodeArg
Onnx/Op/Schema/Kernel
ORTValue/Tensor
GraphTransformer
Per-op Gradient Building
Performance Investigation

Intermediate

ExecutionProvider
IOBinding/dlpack
PyBind
Gradient Graph Building
CUDA Programming

Advanced

Execution Engine
- SessionState
- ExecutionFrame
Memory
Distributed Training

Domain Experts

Performance optimization for CUDA kernels
Hyper-parameter tuning