Welcome to AMD MIGraphX’s documentation!

Contents:

• Python User Guide
  • Python Reference
    • shape
    • argument
    • target
    • program
    • parse_onnx
    • parse_tf
    • load
    • save
• C++ User Guide
  • C++ Reference
    • shape
    • argument
    • target
    • program
    • quantize
    • parse_onnx
    • load
    • save
• MIGraphX Driver
  • read
  • compile
  • run
  • perf
  • verify
  • roctx
• Contributor Guide
  • MIGraphX Fundamentals
    • Location of the Examples
    • Adding Two Literals
    • Using Parameters
    • Handling Tensor Data
    • Importing From ONNX
  • Data types
    • shape
    • literal
    • argument
    • raw_data
    • tensor_view
  • Operators
    • operation
    • operators
  • Program
    • instruction
    • instruction_ref
    • program
    • parse_onnx
    • parse_tf
    • onnx_options
    • tf_options
  • Targets
    • target
    • gpu::target
    • cpu::target
  • Quantization
    • quantize_fp16
    • quantize_int8
  • Passes
    • pass
    • dead_code_elimination
    • eliminate_common_subexpression
    • eliminate_concat
    • eliminate_contiguous
    • eliminate_identity
    • eliminate_pad
    • propagate_constant
    • rewrite_batchnorm
    • rewrite_rnn
    • schedule
    • simplify_algebra
    • simplify_reshapes
  • Matchers
    • Introduction
    • Arguments
    • Binding
    • Finding matches
    • Creating matchers
  • Tools
    • roctx.py

Indices and tables

• Index

• Module Index

• Search Page