tensorrtx: Error Retinafece engine creation

Env

  • Xavier NX
  • Ubuntu18.04
  • Cuda version = 11.4
  • TensorRT version = 8.4

About this repo

  • which branch/tag/commit are you using? latest
  • which model? retinaface?

Your problem

  • what is your command? e.g. sudo ./retina_r50 -s

  • what’s your output? [07/11/2023-20:29:21] [W] [TRT] The implicit batch dimension mode has been deprecated. Please create the network with NetworkDefinitionCreationFlag::kEXPLICIT_BATCH flag whenever possible. Loading weights: …/retinaface.wts Your platform support int8: 1 Building engine, please wait for a while… reading calib cache: r50_int8calib.table [07/11/2023-20:29:38] [E] [TRT] 1: Unexpected exception _Map_base::at Build engine successfully! retina_r50: /home/sensormatic/Neo/tensorrtx/retinaface/retina_r50.cpp:251: void APIToModel(unsigned int, nvinfer1::IHostMemory**): Assertion `engine != nullptr’ failed. Aborted

  • what output do you expect? i was expecting trt engine file.

About this issue

  • Original URL
  • State: closed
  • Created a year ago
  • Comments: 21 (3 by maintainers)

Most upvoted comments

Same thing is happening to me right now. I remember that it worked for me back then so I guess the issue is with the dependencies.