Trt10 (#1552)

* The v5-cls model supports TensorRT10

* The v5-cls model supports TensorRT10 Python API

* add YOLOv5-cls readme

* pre-commit and modify trtx download branch

* pre-commit

yolov5/yolov5_trt10/CMakeLists.txt

@@ -0,0 +1,39 @@
+cmake_minimum_required(VERSION 3.10)
+
+project(yolov5)
+
+add_definitions(-std=c++11)
+add_definitions(-DAPI_EXPORTS)
+option(CUDA_USE_STATIC_CUDA_RUNTIME OFF)
+set(CMAKE_CXX_STANDARD 11)
+set(CMAKE_BUILD_TYPE Debug)
+
+# TODO(Call for PR): make cmake compatible with Windows
+set(CMAKE_CUDA_COMPILER /usr/local/cuda/bin/nvcc)
+enable_language(CUDA)
+
+# include and link dirs of cuda and tensorrt, you need adapt them if yours are different
+# cuda
+include_directories(/usr/local/cuda/include)
+link_directories(/usr/local/cuda/lib64)
+
+# tensorrt
+# TODO(Call for PR): make TRT path configurable from command line
+include_directories(/workspace/shared/TensorRT-10.2.0.19/include/)
+link_directories(/workspace/shared/TensorRT-10.2.0.19/lib/)
+
+include_directories(${PROJECT_SOURCE_DIR}/src/)
+include_directories(${PROJECT_SOURCE_DIR}/plugin/)
+file(GLOB_RECURSE SRCS ${PROJECT_SOURCE_DIR}/src/*.cpp ${PROJECT_SOURCE_DIR}/src/*.cu)
+file(GLOB_RECURSE PLUGIN_SRCS ${PROJECT_SOURCE_DIR}/plugin/*.cu)
+
+add_library(myplugins SHARED ${PLUGIN_SRCS})
+target_link_libraries(myplugins nvinfer cudart)
+
+find_package(OpenCV)
+include_directories(${OpenCV_INCLUDE_DIRS})
+
+add_executable(yolov5_cls yolov5_cls.cpp ${SRCS})
+target_link_libraries(yolov5_cls nvinfer)
+target_link_libraries(yolov5_cls cudart)
+target_link_libraries(yolov5_cls ${OpenCV_LIBS})

yolov5/yolov5_trt10/README.md

@@ -0,0 +1,72 @@
+## Introduce
+
+Yolov5 model supports TensorRT-10.
+
+## Environment
+
+CUDA: 11.8
+CUDNN: 8.9.1.23
+TensorRT: TensorRT-10.2.0.19
+
+## Support
+
+* [x] YOLOv5-cls support FP32/FP16/INT8 and Python/C++ API
+
+## Config
+
+* Choose the YOLOv5 sub-model n/s/m/l/x/n6/s6/m6/l6/x6 from command line arguments.
+* Other configs please check [src/config.h](src/config.h)
+
+## Build and Run
+
+1. generate .wts from pytorch with .pt, or download .wts from model zoo
+
+```shell
+git clone -b v7.0 https://github.com/ultralytics/yolov5.git
+git clone -b trt10 https://github.com/wang-xinyu/tensorrtx.git
+cd yolov5/
+wget https://github.com/ultralytics/yolov5/releases/download/v7.0/yolov5n.pt
+cp [PATH-TO-TENSORRTX]/yolov5/gen_wts.py .
+python gen_wts.py -w yolov5n.pt -o yolov5n.wts
+# A file 'yolov5n.wts' will be generated.
+```
+
+2. build tensorrtx/yolov5/yolov5_trt10 and run
+
+#### Classification
+
+```shell
+cd [PATH-TO-TENSORRTX]/yolov5/yolov5_trt10
+# Update kNumClass in src/config.h if your model is trained on custom dataset
+mkdir build
+cd build
+cp [PATH-TO-ultralytics-yolov5]/yolov5s.wts .
+cmake ..
+make
+
+# Download ImageNet labels
+wget https://github.com/joannzhang00/ImageNet-dataset-classes-labels/blob/main/imagenet_classes.txt
+
+# Build and serialize TensorRT engine
+./yolov5_cls -s yolov5n-cls.wts yolov5n-cls.engine [n/s/m/l/x]
+
+# Run inference
+./yolov5_cls -d yolov5n-cls.engine ../../images
+# The results are displayed in the console
+```
+
+3. Optional, load and run the tensorrt model in Python
+```shell
+// Install python-tensorrt, pycuda, etc.
+// Ensure the yolov5n-cls.engine
+python yolov5_cls_trt.py
+```
+
+## INT8 Quantization
+1. Prepare calibration images, you can randomly select 1000s images from your train set. For coco, you can also download my calibration images `coco_calib` from [GoogleDrive](https://drive.google.com/drive/folders/1s7jE9DtOngZMzJC1uL307J2MiaGwdRSI?usp=sharing) or [BaiduPan](https://pan.baidu.com/s/1GOm_-JobpyLMAqZWCDUhKg) pwd: a9wh
+2. unzip it in yolov5_trt10/build
+3. set the macro `USE_INT8` in src/config.h and make again
+4. serialize the model and test
+
+## More Information
+See the readme in [home page.](https://github.com/wang-xinyu/tensorrtx)