# runtime_custom_op

**Repository Path**: zhaowujie/runtime_custom_op

## Basic Information

- **Project Name**: runtime_custom_op
- **Description**: onnxruntime custom op library
- **Primary Language**: Unknown
- **License**: Apache-2.0
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-01-17
- **Last Updated**: 2025-01-24

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# ONNXRuntime custom operator implementation sample

## motivation
[ONNX](https://github.com/onnx/onnx) model can have custom operator.
But tools somewhat like [ONNX Simplifier](https://github.com/daquexian/onnx-simplifier) needs inference by [ONNXRuntime](https://github.com/microsoft/onnxruntime/).
So we need implementation of custom operator for onnxruntime.

## samples
There are simple FMA(fused multiply-add) custom operator samples.

* [pure Python sample](fma_pure_python)
* [C++ implementation sample](fma_use_t)
* [C++ implementation multi-domain sample](fma_multi_domain)
* [CUDA implementation multi-domain sample](fma_multi_with_cuda)