1 Execution Engine

Sherlock редактировал(а) эту страницу 2021-03-12 13:58:08 -08:00

Содержание

• InferenceSession::Run()
• SequentialExecutor::Execute()
• Advance: GraphPartitioner

InferenceSession::Run()

• Read RunOptions and understand the options (run_option.h)

SequentialExecutor::Execute()

• What's the purpose of ExecutionFrame? (execution_frame.h)

  • How is one nodes output passed in as another node's input?
  • What happens when we call context->Output() inside an op kernel?
  • How are feeds and fetches stored in ExecutionFrame?

• How is the execution order determined? (graph_viewer.cc)

  • Default execution order uses Graph::ReverseDFS() to generated topological sort
  • Priority-based execution order uses Graph::KahnsTopologicalSort with per-node priority

• How is each node's kernel invoked ?

• How does ORT guarantees all the cuda kernel is completed before session.run return?

Advance: GraphPartitioner

• How each node is determined to be place on which execution provider? (graph_partitioner.h)

Please use the learning roadmap on the home wiki page for building general understanding of ORT.

Navigation by topic

Upcoming Release Roadmap

Glossary

Development

• Github Development
• Tips & Tricks
• Logging
• Leverage Existing Artifacts

Common Tasks

• Run a model
• Write a kernel
• Investigate performance shortcomings
• Testing
• Update documentation

Dependencies

• ONNX

Core Architecture

• Execution Engine
• Execution Provider
• Graph
• Memory Management
• Ops and Kernels
• User Interface

Feature Details

• Contrib Ops
• CUDA Programming

Inferencing

Training

• Model Training Domain Knowledge
• Distributed Training