I am trying to use Tensorflow on a NVIDIA GRID K1. The last version of supported Cuda is 10.x.x. So, I installed Cuda 10.1 and according to here https://www.tensorflow.org/install/source_windows?hl=fr I installed tensorflow-gpu 2.3.0 and Python 3.8.10.
By running this : tf.config.list_physical_devices('GPU') in Python I get this :
tensorflow/core/common_runtime/gpu/gpu_device.cc:1812] Ignoring visible gpu device (device: 0, name: GR ID K1, pci bus id: 0000:05:00.0, compute capability: 3.0) with Cuda compute capability 3.0. The minimum required Cuda capability is 3.5.
I've found some fix on Google as The minimum required Cuda capability is 3.5 but I'm using tensorflow-gpu and I do not find any file "configure" and where to put this : TF_UNOFFICIAL_SETTING=1.
On other forum, I have found that in 2015 a patch was develop but nothing since this time about this patch.
Does anyone have an idea ?
