tensorrtx
tensorrtx copied to clipboard
虽然最终得到了engine模型,但是报了这个错误,我是否应该忽视
Env
- GPU3090ti
- Win11
- Cuda version:11.8
- TensorRT version:8.6.1.6
About this repo
- tag: yolov5s v6.0
- yolov5
My problem
-
my command: yolov5.exe -s yolo5_wc_10_5x5.wts yolov5_4x5_bt256.engine s
-
what's your output? [06/04/2024-10:20:35] [W] [TRT] The implicit batch dimension mode has been deprecated. Please create the network with NetworkDefinitionCreationFlag::kEXPLICIT_BATCH flag whenever possible. Loading weights: yolo5_wc_10_5x5.wts Building engine, please wait for a while... [06/04/2024-10:27:06] [W] [TRT] TensorRT encountered issues when converting weights between types and that could affect accuracy. [06/04/2024-10:27:06] [W] [TRT] If this is not the desired behavior, please modify the weights or retrain with regularization to adjust the magnitude of the weights. [06/04/2024-10:27:06] [W] [TRT] Check verbose logs for the list of affected weights. [06/04/2024-10:27:06] [W] [TRT] - 50 weights are affected by this issue: Detected subnormal FP16 values. [06/04/2024-10:27:06] [W] [TRT] - 2 weights are affected by this issue: Detected values less than smallest positive FP16 subnormal value and converted them to the FP16 minimum subnormalized value. Build engine successfully! [06/04/2024-10:27:06] [E] [TRT] 3: [builder.cpp::nvinfer1::builder::Builder::~Builder::341] Error Code 3: API Usage Error (Parameter check failed at: builder.cpp::nvinfer1::builder::Builder::~Builder::341, condition: mObjectCounter.use_count() == 1. Destroying a builder object before destroying objects it created leads to undefined behavior. )
-
what output do you expect? why appear"[06/04/2024-10:27:06] [E] [TRT] 3: [builder.cpp::nvinfer1::builder::Builder::~Builder::341] Error Code 3: API Usage Error (Parameter check failed at: builder.cpp::nvinfer1::builder::Builder::~Builder::341, condition: mObjectCounter.use_count() == 1. Destroying a builder object before destroying objects it created leads to undefined behavior. )",shouild i ignore or take some meatures?