tensorRT_Pro icon indicating copy to clipboard operation
tensorRT_Pro copied to clipboard

C++ library based on tensorrt integration

Results 98 tensorRT_Pro issues
Sort by recently updated
recently updated
newest added

自己训练的两个模型预测结果准确度大幅下降,yolo预训练模型没有问题,请问可能是什么方面的问题?

运行时报错信息:[2023-02-07 11:22:22][error][trt_builder.cpp:30]:NVInfer: C:\source\rtSafe\safeRuntime.cpp (32) - Cuda Error in nvinfer1::internal::DefaultAllocator::free: 1 (invalid argument) 经检查,这个错误是在执行:engine_ = shared_ptr(runtime_->deserializeCudaEngine(pdata, size, nullptr), destroy_nvidia_pointer);时报出的。(反序列化引擎) 相同的cuda、cudnn、tensorrt版本,在linux中正常运行,但迁移到windows中时,始终报这个错误,无法加载引擎。 可以确定序列化前后tensorrt版本完全一致。 奇怪的是,windows中相同环境,只把tensorrt换成tensorrt7,即可正常运行,还试了其他tensorrt版本,都报上面的错误。 环境: VS2019 cuda11.3 cudnn8.0.4 tensorrt8.2.2.1

Hi [2022-12-08 17:36:09][error][trt_builder.cpp:30]:NVInfer: 9: [graphShapeAnalyzer.cpp::nvinfer1::builder::`anonymous-namespace'::ShapeNodeRemover::throwIfError::1306] Error Code 9: Internal Error (Reshape_1087: reshape changes volume ) [2022-12-08 17:36:09][error][trt_builder.cpp:30]:NVInfer: 9: [graphShapeAnalyzer.cpp::nvinfer1::builder::`anonymous-namespace'::ShapeNodeRemover::throwIfError::1306] Error Code 9: Internal Error (Reshape_1087: reshape changes volume ) [2022-12-08...

https://github.com/ultralytics/yolov5 https://github.com/WongKinYiu/yolov7 Are they the same preprocess and decode ?

非常感兴趣作者的dockerfile文件,可以开源提供嘛?

## 编译环境 * CUDA 11.7 * CUDNN 8.5 * TensorRT 8.4 * clang 14.0.6 ## 错误信息 ``` tensorRT_Pro/src/tensorRT/common/trt_tensor.hpp:122:53: fatal error: non-constant-expression cannot be narrowed frod long' to 'int' in initializer...

对分割模型进行推理的时候 用yolo 对intput0输出画框 输出框的一个目标有多个框的情况 ![image](https://user-images.githubusercontent.com/51087601/198211206-a27a9070-d484-4597-9adb-5b7871a0f98b.png) ![image](https://user-images.githubusercontent.com/51087601/198211292-4b6c9e6a-2ffd-4aa2-ac2b-078813012f35.png) 是后处理用问题吗? ![image](https://user-images.githubusercontent.com/51087601/198211515-5574f72b-aced-4520-a1eb-8faa7d7f640e.png)

C:\Users\17243\Desktop\tensorRT_Pro-main\example-python>python test_yolov5.py [2022-09-30 08:54:25][error][trt_builder.cpp:30]:NVInfer: 1: [stdArchiveReader.cpp::nvinfer1::rt::StdArchiveReader::StdArchiveReader::34] Error Code 1: Serialization (Serialization assertion safeVersionRead == safeSerializationVersion failed.Version tag does not match. Note: Current Version: 43, Serialized Engine Version: 0) [2022-09-30 08:54:25][error][trt_builder.cpp:30]:NVInfer:...

C:\Users\17243\Desktop\tensorRT_Pro-main\example-python>python test_yolov5.py Traceback (most recent call last): File "C:\Users\17243\Desktop\tensorRT_Pro-main\example-python\test_yolov5.py", line 4, in import pytrt as tp File "C:\Users\17243\Desktop\tensorRT_Pro-main\example-python\pytrt\__init__.py", line 335, in from .libpytrtc import * ModuleNotFoundError: No module named 'pytrt.libpytrtc'