I saw in the description that it is possible to use the Tensor cores of the GPU
However, this option does not appear on both RTX2080 and RTX4090
I would like to know if it is not developed yet or if there is something wrong with my configuration.
Windows DirectML will use the Nvidia GPU
Although TensorRT might be available, I think it's mostly for Linux
CUDA is also availaebl in ONNX Runtime for Windows... I may look into that
I saw in the description that it is possible to use the Tensor cores of the GPU However, this option does not appear on both RTX2080 and RTX4090 I would like to know if it is not developed yet or if there is something wrong with my configuration.