search

microsoft / DirectML

DirectML is a high-performance, hardware-accelerated DirectX 12 library for machine learning. DirectML provides GPU acceleration for common machine learning tasks across a broad range of supported hardware and drivers, including all DirectX 12-capable GPUs from vendors such as AMD, Intel, NVIDIA, and Qualcomm.
MIT License
2.21k stars 294 forks source link

Hi, please help on aten::softplus.out #586

Closed DeruVN closed 4 months ago

DeruVN commented 5 months ago

Hi, please help on aten::softplus.out

The operator 'aten::softplus.out' is not currently supported on the DML backend and will fall back to run on the CPU.The operator 'aten::softplus.out' is not currently supported on the DML backend and will fall back to run on the CPU.

DeruVN commented 5 months ago

I'm trying run torch-directml, AMD RX 6800, for ZoeDepth model https://github.com/isl-org/ZoeDepth

joshjkim commented 4 months ago

Hi @DeruVN , thanks for your feedback. We're adding support for the softplus operator in our upcoming torch-directml build releasing soon. With our latest commitment to expand PyTorch support on DirectML, you can expect to see more regular updates from us to address any community feedback and support the latest and greatest PyTorch features on DirectML. Stay tuned!

joshjkim commented 4 months ago

@DeruVN Here's the link to our latest torch-directml 0.2.2.dev240614 build with DML support for softplus.out. Please run pip install torch-directml --upgrade to update