1 Topic by badhomaks (edited by badhomaks 17-03-2024 23:59:36)

  • Thanks: 1

Topic: RIFE: trtexec execution fails

I've attached the logs to the post as well (not sure why not showing up on the post so I'll just paste it below). Not quite sure why it's giving me issues to set up. The optimization option also crashes but at least the rife v13 and below still worked, but the v15 just outright refuses to work. Any tips? I've tried reinstalling but that didn't do anything.




[03/17/2024-16:38:29] [W] Could not read timing cache from: C:\Users\Artur\AppData\Roaming\SVP4\cache\Program Files (x86)/SVP 4/rife\models\rife\rife_v4.15.onnx.min64x64_opt1920x1088_max2560x1440_fp16_no-tf32_trt-8502_cudnn_I-fp16_O-fp16_NVIDIA-GeForce-RTX-3080_74a76e08.engine.cache. A new timing cache will be generated and written.
[03/17/2024-16:38:29] [E] Error[4]: [network.cpp::nvinfer1::Network::validate::3100] Error Code 4: Internal Error (input: for dimension number 1 in profile 0 does not match network definition (got min=11, opt=11, max=11), expected min=opt=max=7).)
[03/17/2024-16:38:29] [E] Error[2]: [builder.cpp::nvinfer1::builder::Builder::buildSerializedNetwork::751] Error Code 2: Internal Error (Assertion engine != nullptr failed. )
[03/17/2024-16:38:29] [E] Engine could not be created from network
[03/17/2024-16:38:29] [E] Building engine failed
[03/17/2024-16:38:29] [E] Failed to create engine from model or file.
[03/17/2024-16:38:29] [E] Engine set up failed
&&&& FAILED TensorRT.trtexec [TensorRT v8501] # C:/Program Files (x86)/SVP 4/rife\vsmlrt-cuda\trtexec --onnx=C:/Program Files (x86)/SVP 4/rife\models\rife\rife_v4.15.onnx --timingCacheFile=C:\Users\Artur\AppData\Roaming\SVP4\cache\Program Files (x86)/SVP 4/rife\models\rife\rife_v4.15.onnx.min64x64_opt1920x1088_max2560x1440_fp16_no-tf32_trt-8502_cudnn_I-fp16_O-fp16_NVIDIA-GeForce-RTX-3080_74a76e08.engine.cache --device=0 --saveEngine=C:\Users\Artur\AppData\Roaming\SVP4\cache\Program Files (x86)/SVP 4/rife\models\rife\rife_v4.15.onnx.min64x64_opt1920x1088_max2560x1440_fp16_no-tf32_trt-8502_cudnn_I-fp16_O-fp16_NVIDIA-GeForce-RTX-3080_74a76e08.engine --minShapes=input:1x11x64x64 --optShapes=input:1x11x1088x1920 --maxShapes=input:1x11x1440x2560 --fp16 --tacticSources=-CUBLAS,-CUBLAS_LT --useCudaGraph --noDataTransfers --noTF32 --inputIOFormats=fp16:chw --outputIOFormats=fp16:chw --preview=+fasterDynamicShapes0805 --layerPrecisions=*:fp16 --layerOutputTypes=*:fp16 --precisionConstraints=obey

2 Reply by flowreen91 (edited by flowreen91 18-03-2024 06:59:36)

  • Thanks: 48
  • Thanks for the post: 1

Re: RIFE: trtexec execution fails

badhomaks wrote:

Internal Error (input: for dimension number 1 in profile 0 does not match network definition (got min=11, opt=11, max=11), expected min=opt=max=7).)

https://www.svp-team.com/forum/viewtopi … 125#p84125
https://www.svp-team.com/forum/viewtopi … 126#p84126

3 Reply by badhomaks (edited by badhomaks 21-03-2024 01:27:28)

  • Thanks: 1

Re: RIFE: trtexec execution fails

Okay, so it appears to not cause that error but after starting inference, it only lasts a few seconds before closing the cmd window (screenshot of the window right before it closes).https://i.imgur.com/TjXeu1n.png

I vaguely remember that they usually take a few minutes and did that before I updated SVP a year or so ago and since then the optimization passes became very short. The video at least seems to run. Using a 3090 if it makes a difference.

  • Thanks: 48

Re: RIFE: trtexec execution fails

badhomaks wrote:

The video at least seems to run.

Good to know u fixed it.
Always try to run RIFE with "Performance Boost" Enabled so it will reuse the generated cache file. They probably improved something and it takes less and less time to generate.
Never use the ensemble onnx cause it takes an overwhelming amount of processing power compared to the others.