Fabian Schuetze

Setting max_gpu_fraction as in Tensorflow backend

2

This PR permits setting the a max_gpu_fraction for the pytorch backend. Pytorch allows setting the max gpu fraction through the [`CUDACachingAllocator`](https://github.com/pytorch/pytorch/blob/main/c10/cuda/CUDACachingAllocator.h) . The user of the pytorch_backend can set the...

[egl OffscreenContext:] Bindless Texture Not supported

8

Thanks for continuing the work on GDR. I enjoyed working with the initial version a year ago and look forward to trying the revised version. When I start the training,...

Thanks for the wonderful paper - it was a pleasure to read! Could you kindly elaborate a bit more on the COCO training details? In particular, I was wondering about...

[HVX] Blur example fails with: Failed to write shared object to 'libhalide_shared_runtime.so'

1

I am trying to execute the blur example on an Snapdragon SM8250 (Samung Galaxy Tablet S7), but running the blur example fails with the error: ``` Aborted ``` `Adb Logcat...

[HEXAGON] Permit hvx_128 instruction in Blur Target

5

I noticed that the variable `HL_TARGET` is not used in the generator when building the blur app for a Hexagon device. Instead, the target is always the generic `arm-64-android`. The...

Thanks for the wonderful library. Apologies if this seems to be a silly question: How can I benchmark gemms on a android target? In line with the docs for [test](https://arm-software.github.io/ComputeLibrary/latest/tests.xhtml)...