search

shouxieai / tensorRT_Pro

C++ library based on tensorrt integration
MIT License
2.62k stars 549 forks source link

yolov5 7.0 warning: build done 65s; fatal error Assert failed, this->blobsNameMapper_.find(name) != this->blobsNameMapper_.end() #205

Open HaronCHou opened 1 year ago

HaronCHou commented 1 year ago

yolov5 v7.0 遇到的问题如下：

  1. build时间过长
  2. core dumped 原因未知
(base) root@f3cd9dc49bd1:/datav/projects/tensorRT_Pro/simple_yolo/workspace# /datav/projects/tensorRT_Pro/simple_yolo/workspace/pro
===================== test YoloV5 FP32 yolov5s2 ==================================
[info][simple_yolo.cu:2216]:Compile FP32 yolov5s2_dynamic.onnx.
[warn][simple_yolo.cu:1116]:NVInfer: onnx2trt_utils.cpp:364: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[warn][simple_yolo.cu:1116]:NVInfer: Output type must be INT32 for shape outputs
[warn][simple_yolo.cu:1116]:NVInfer: Output type must be INT32 for shape outputs
[warn][simple_yolo.cu:1116]:NVInfer: Output type must be INT32 for shape outputs
[info][simple_yolo.cu:2277]:Input shape is -1 x 3 x 640 x 640
[info][simple_yolo.cu:2278]:Set max batch size = 16
[info][simple_yolo.cu:2279]:Set max workspace size = 1024.00 MB
[info][simple_yolo.cu:2282]:Network has 1 inputs:
[info][simple_yolo.cu:2288]:      0.[images] shape is -1 x 3 x 640 x 640
[info][simple_yolo.cu:2294]:Network has 1 outputs:
[info][simple_yolo.cu:2299]:      0.[output0] shape is 1 x 25200 x 85
[info][simple_yolo.cu:2303]:Network has 329 layers
[info][simple_yolo.cu:2319]:Building engine...
[warn][simple_yolo.cu:1116]:NVInfer: TensorRT was linked against cuBLAS/cuBLAS LT 11.5.1 but loaded cuBLAS/cuBLAS LT 11.4.1
[warn][simple_yolo.cu:1116]:NVInfer: Detected invalid timing cache, setup a local cache instead
[warn][simple_yolo.cu:1116]:NVInfer: Max value of this profile is not valid
[warn][simple_yolo.cu:1116]:NVInfer: TensorRT was linked against cuBLAS/cuBLAS LT 11.5.1 but loaded cuBLAS/cuBLAS LT 11.4.1
[info][simple_yolo.cu:2340]:Build done 65426 ms !
[warn][simple_yolo.cu:1116]:NVInfer: TensorRT was linked against cuBLAS/cuBLAS LT 11.5.1 but loaded cuBLAS/cuBLAS LT 11.4.1
[warn][simple_yolo.cu:1116]:NVInfer: TensorRT was linked against cuBLAS/cuBLAS LT 11.5.1 but loaded cuBLAS/cuBLAS LT 11.4.1
[info][simple_yolo.cu:1263]:Infer 0x7f42f8000c90 detail
[info][simple_yolo.cu:1264]:    Max Batch Size: 16
[info][simple_yolo.cu:1265]:    Inputs: 1
[info][simple_yolo.cu:1269]:            0.images : shape {16 x 3 x 640 x 640}
[info][simple_yolo.cu:1272]:    Outputs: 1
[info][simple_yolo.cu:1276]:            0.output0 : shape {16 x 25200 x 85}
[fatal][simple_yolo.cu:1506]:Assert failed, this->blobsNameMapper_.find(name) != this->blobsNameMapper_.end()
Aborted (core dumped)