Skip to main content
Ctrl+K
onnx-light 0.1.31 documentation - Home onnx-light 0.1.31 documentation - Home
  • Quick tour
  • Start
  • Design
  • API
  • ByOp
  • How-To
  • Next Steps
  • Miscellaneous
  • GitHub
  • Quick tour
  • Start
  • Design
  • API
  • ByOp
  • How-To
  • Next Steps
  • Miscellaneous
  • GitHub

Section Navigation

  • Introduction to ONNX
    • ONNX Concepts
    • ONNX with Python
  • Proto
    • Goals
    • Detailed Differences between onnx and onnx_light
    • Protobuf format applied to ONNX
    • ORT flatbuffer format: parallelization and alignment
    • ModelProto creation and no-copy ownership
    • Complex loading and saving scenarios
    • Fuzzing
  • Expressions
  • Compute
    • Value-as-shape propagation
    • Symbolic-dimension constraint mechanism
    • Sequences, maps and subgraphs
    • Shape-inference events (ShapeEvent)
    • Shape-inference coverage
  • Patterns
    • GraphBuilder
    • Pattern optimization
  • Gradient
  • Runtime
    • Buffer-reuse arenas
    • Backend test-case coverage
    • Runtime test coverage (onnxruntime and shape_inference)
  • Tuning
    • Processor-aware kernel tuning
    • Backend calibration profile storage
  • Technical
    • Thread-pool dispatch in onnx-light, OpenMP, and ONNX Runtime
    • Overlapping ONNX loading and prepacking
  • Design
  • Technical

Technical#

These pages describe implementation details and compare the execution strategies used by onnx-light and other runtimes.

  • Thread-pool dispatch in onnx-light, OpenMP, and ONNX Runtime
  • Overlapping ONNX loading and prepacking

previous

Backend calibration profile storage

next

Thread-pool dispatch in onnx-light, OpenMP, and ONNX Runtime
Show Source

Created using Sphinx 9.1.0.

Built with the PyData Sphinx Theme 0.22.0.