Michel van Lier comments

Results 20 comments of


                                            Michel van Lier

[QST] Is it possible to exchange data on GPU with OpenCV CUDA?

Thank you for your example! Unfortunately, it is not working. My array is a cv2.CV_8UC3 (see https://gist.github.com/yangcha/38f2fa630e223a8546f9b48ebbb3e61a) Therefore have set as follows: `"typestr": "u1"` But I am getting the following...

chuangmi.plug.hmi206 support

I would also appreciate the support for this plug! @hatebreeder94 did you already found an option to get the chuangmi.plug.hmi206 working with homekit/homebridge?

Unclear usage of training look and training_steps in ActivationAPoZRankPruner (model compression, pruning)

Thanks for your reply and explanation. Could you provide any timeline on the next compression version where this will be fixed? What would be a good workaround in the meantime?...

Wrong kpts in train batch

@Gwencong I already implemented this, but it doesn't help

Attention imporved yolov5 performance

@315386775 how dit you implemented this: "_add conv layer before and later with FPN_." ?

fix & speedup show_gui

Any update on this?

Deep Learning model in TensorRT with SPARSE layers not accelerating inference speed

Thanks for the quick response I have made the links accessible now, but just to be sure here are the links: - [ONNX sparse batch-size = 1](https://drive.google.com/file/d/1IyHqK9mI3gYf1aDIjmeE7OXIGSVKWHwW/view?usp=sharing0) - [ONNX non-sparse-...

Deep Learning model in TensorRT with SPARSE layers not accelerating inference speed

@lix19937 @nvpohanh I have open the acces to the models, could you have a look for me?

Deep Learning model in TensorRT with SPARSE layers not accelerating inference speed

@nvpohanh, thanks for checking. Do you have links to documentation with more details? (I couldn't find it.) What are the technical details of why _"input/output channels of the convs are...

Deep Learning model in TensorRT with SPARSE layers not accelerating inference speed

@nvpohanh So I did some tests with the backbone of yolov8l (first 20 layers), changing the backbone of the model to the rules you mentioned (as you can see below)...