jensdraht1999

Tips: Do NOT USE IDLE. There is really something broke with it. Use Cmd or Powershell or something else. It really worked with the example, if run through cmd. There's...

Updated with 12.3 cuda and made all the steps again. Now it seems to work better, but I'm not sure, if the example provided is suppossed to have exactly this...

@gs-olive I do it like this: The code I have saved in a file called 1.py. The content: ``` import torch import torchvision model = models.resnet18(pretrained=True).half().eval().to("cuda") inputs = [torch.randn((1, 3,...

This will happen in Powershell if import is done: PS C:\Users\{UserName}> python Python 3.10.11 (tags/v3.10.11:7d4cc5a, Apr 5 2023, 00:38:17) [MSC v.1929 64 bit (AMD64)] on win32 Type "help", "copyright", "credits"...

WAIT A MINUTE: Could it be, that regular GPUs like RTX 3060 LAPTOP are not supported by Tensorrt at all?

It has worked with the code you have provided. Thank you. @HolyWu Here is the what is the output: ``` Microsoft Windows [Version 10.0.19045.3930] (c) Microsoft Corporation. Alle Rechte vorbehalten....

@HolyWu Just wanted to ask you, did it work for you, without deleting "C:\Users\{UserName}\AppData\Local\Programs\Python\Python310\Lib\site-packages\nvidia\cudnn" folder?

I wonder if this has something to do with your respective caches, which is very high on 12900k (14 + 30Mb) so bigger cache cpus will gain more out of...

@stickz any Updates?

> Some really basic initial testing to begin with has before/after within a margin of error on my 5900X so whatever issue was showing up before seems to have disappeared....