Deep Learning API and Server in C++14 support for PyTorch,TensorRT, Dlib, NCNN, Tensorflow, XGBoost and TSNE

InsightFace REST API for easy deployment of face recognition services with TensorRT in Docker.

针对pytorch模型的自动化模型结构分析和修改工具集，包含自动分析模型结构的模型压缩算法库

End2EndPerception deployment solution based on vision sparse transformer paradigm is open sourced.

Based on tensorrt v8.0+, deploy detection, pose, segment, tracking of YOLO11 with C++ and python api.

Based on tensorrt v8.0+, deploy detect, pose, segment, tracking of YOLOv8 with C++ and python api.

Yolov5 TensorRT Implementations

Using TensorRT for Inference Model Deployment.

this is a tensorrt version unet, inspired by tensorrtx

VitPose without MMCV dependencies

Advanced inference pipeline using NVIDIA Triton Inference Server for CRAFT Text detection (Pytorch), included converter from Pytorch -> ONNX -> TensorRT, Inference pipelines (TensorRT, Triton server - multi-format). Supported model format for Triton inference: TensorRT engine, Torchscript, ONNX

Base on tensorrt version 8.2.4, compare inference speed for different tensorrt api.

The real-time Instance Segmentation Algorithm SparseInst running on TensoRT and ONNX

Convert yolo models to ONNX, TensorRT add NMSBatched.

Based on TensorRT v8.2, build network for YOLOv5-v5.0 by myself, speed up YOLOv5-v5.0 inferencing

Advance inference performance using TensorRT for CRAFT Text detection. Implemented modules to convert Pytorch -> ONNX -> TensorRT, with dynamic shapes (multi-size input) inference.

tensorrt-toy code

Simple tool for PyTorch >> ONNX >> TensorRT conversion

Dockerized TensorRT inference engine with ONNX model conversion tool and ResNet50 and Ultraface preprocess and postprocess C++ implementation

Export (from Onnx) and Inference TensorRT engine with Python