mahaitao999 Goto Github PK

onnx_trt_cpp

采用TensorRT的C++接口进行ONNX模型转TRT，并进行Inference推理。前处理采用C++版本的OpenCV。

pointers_on_c

《C和指针》一书的代码

pointpillars_multihead_40fps

A REAL-TIME 3D detection network [Pointpillars] compiled by CUDA/TensorRT/C++.

professional-cuda-c-programming

pybind11_cuda

使用pybind11进行Python、C++以及CUDA的混合编程

pybind11_sample

pybind11库的一些简单使用例子

pytorch-cpp-inference

Serving PyTorch 1.0 Models as a Web Server in C++

pytorch_extension_custom_op

PyTorch自定义算子

retinaface-pytorch-onnx-trt

retinaface的训练及优化部署

service-streamer

Boosting your Web Services of Deep Learning Applications.

tensorrt-utils

⚡ Useful scripts when using TensorRT

tensorrt_plugin_template

TensorRT实现Plugin的一些模板化流程

tensorrt_python_api

用TensorRT的Python API去搭建trt网络

tensorrtintegrate

TensorRT ONNX Plugin、Inference、Compile

tiny-tensorrt

a simple, efficient, easy-to-use nvidia TensorRT wrapper for cnn,sopport c++ and python

top-down-approach

myk's learning material of Computer Network: Top Down Approach

trt-samples-for-hackathon-cn

Simple samples for TensorRT programming

upsamplingplugin

A upsampling layer for tenosrrt...

wsgi_example

这是我在实践中发现采用Gunicorn的Web Server时，传输数据比较大时（比如说一张1080x720的图片其解码完的图像矩阵～6M），要明显比Flask默认的Werkzurg慢得多。由于Gunicorn是以命令行的方式启动的，所以调试不太方便，所以我就按照原始的WSGI方式写了个测试代码。

yolov3_dynamic_batch_tensorrt_triton

将Yolov3模型转成可以进行动态Batch的TensorRT推理以及Triton Inference Serving上部署的TensorRT模型

yolov4-triton-tensorrt

This repository deploys YOLOv4 as an optimized TensorRT engine to Triton Inference Server