Sacha
Sacha
When it gets converted to a tensor, the default type is float and the default action in Python will normalise it. However, it won't do this automatically in C++. So,...
That's the mean being subtracted from the image as an unsigned char array, prior to normalisation. That BGR value becomes the zero of the normalised values. I've found the actual...
The inputs are the range of [-104:151, -117:138, -123:132]. They aren't normalised to 1.
You would have to generate the feature separately (face recognition algorithm), taking the detected face as input, to get the 512-D feature embedding.
Where is the documentation that suggests it should be the same? It looks like read() gets you the raw text (i.e. formula).
OK, I think I got it now. When I run the eval in the python script against COCO, it says: [09/29 09:18:20 d2.data.build]: Distribution of instances among all 65 categories:...
@xiaofeng94 I was wondering if you could provide more information about how to change the text encoder. The ./datasets/coco/annotations/open_voc/instances_eval.json contains the classes I want already and this is the file...
> when wsl makes gpu passthrough possible, everything solved Not at all. There's no way to install drivers on WSL.
Intel supports Windows on their new Arc GPUs and still nothing from AMD.
From that repo: "ROCm toolchain for Windows (No public release yet)" Still waiting :( There has been talk of ROCm for Windows for roughly 5 years now with small hints...