microsoft/DirectML issues and pull requests

#362 - [DirectML] Divergent results between CPU and DirectML only on Intel Integrated GPU

Issue - State: open - Opened by apleynes almost 2 years ago - 9 comments
Labels: pytorch-directml, onnx-runtime, hardware

#361 - pytorch-directml produce "[W dml_heap_allocator.cc:97] DML allocator out of memory!"

Issue - State: open - Opened by virtual-feng almost 2 years ago - 7 comments
Labels: pytorch-directml

#360 - torch-directml : torch.div with trunc rounding on int64 fails with RuntimeError

Issue - State: closed - Opened by Theucalyptus almost 2 years ago - 1 comment
Labels: pytorch-directml

#359 - Very low validation and testing accuracy on CNN

Issue - State: open - Opened by AtiqurRahmanAni almost 2 years ago - 3 comments
Labels: pytorch-directml

#356 - Operator 'aten::amax.out' is not currently supported on the DML backend.

Issue - State: open - Opened by rmskmr05 almost 2 years ago - 2 comments
Labels: pytorch-directml

#355 - nn-morse and torch-directml memory leak?

Issue - State: open - Opened by moozoo64 almost 2 years ago - 7 comments
Labels: pytorch-directml

#353 - [Suguestion] Would you consider to open source `torch_directml` and `torch_directml_native`?

Issue - State: open - Opened by HowardZorn almost 2 years ago - 1 comment
Labels: pytorch-directml

#352 - [pytorch-directml] As run a GNN model, meet "Segmentation fault".

Issue - State: closed - Opened by Looong01 almost 2 years ago - 5 comments
Labels: pytorch-directml

#345 - pytorch-directml not working with Stable Diffusion

Issue - State: closed - Opened by Amblyopius almost 2 years ago - 10 comments
Labels: pytorch-directml

#344 - Bump certifi from 2019.11.28 to 2022.12.7 in /TensorFlow/TF1/squeezenet/src

Pull Request - State: closed - Opened by dependabot[bot] almost 2 years ago - 1 comment
Labels: dependencies

#334 - As run a GNN model, meet fatal error

Issue - State: closed - Opened by Looong01 almost 2 years ago - 4 comments
Labels: pytorch-directml

#332 - Bump pillow from 7.0.0 to 9.3.0 in /TensorFlow/TF1/squeezenet/src

Pull Request - State: open - Opened by dependabot[bot] almost 2 years ago
Labels: dependencies

#331 - Bump tensorflow from 1.15.2 to 2.9.3 in /TensorFlow/TF1/squeezenet/src

Pull Request - State: closed - Opened by dependabot[bot] almost 2 years ago - 1 comment
Labels: dependencies

#329 - TensorFlow 2.10 device enumeration failed on Intel(R) HD Graphics 5500

Issue - State: closed - Opened by sabbir2609 almost 2 years ago - 5 comments
Labels: tensorflow

#323 - Pytorch-directml Python 3.9/3.10/3.11 support :)

Issue - State: closed - Opened by Djdefrag about 2 years ago - 7 comments
Labels: pytorch-directml

#323 - Pytorch-directml Python 3.9/3.10/3.11 support :)

Issue - State: closed - Opened by Djdefrag about 2 years ago - 7 comments
Labels: pytorch-directml

#321 - Unsupported operator aten::reflection_pad2d

Issue - State: closed - Opened by Djdefrag about 2 years ago - 3 comments
Labels: pytorch-directml

#305 - XMX-accelerated operations avaliable for Intel Arc GPUs?

Issue - State: open - Opened by tedliosu about 2 years ago - 2 comments

#305 - XMX-accelerated operations avaliable for Intel Arc GPUs?

Issue - State: open - Opened by tedliosu about 2 years ago - 2 comments

#305 - XMX-accelerated operations avaliable for Intel Arc GPUs?

Issue - State: open - Opened by tedliosu about 2 years ago - 2 comments

#301 - Bug when frame buffering

Issue - State: closed - Opened by tom-huntington about 2 years ago - 1 comment

#298 - Bump protobuf from 3.11.3 to 3.18.3 in /TensorFlow/TF1/squeezenet/src

Pull Request - State: open - Opened by dependabot[bot] about 2 years ago
Labels: dependencies

#297 - Bump urllib3 from 1.25.8 to 1.26.5 in /TensorFlow/TF1/squeezenet/src

Pull Request - State: closed - Opened by dependabot[bot] about 2 years ago - 1 comment
Labels: dependencies

#290 - module 'torch' has no attribute 'dml'

Issue - State: closed - Opened by gkiler about 2 years ago - 1 comment

#282 - question about expected performance of quantized operations

Issue - State: open - Opened by carsonswope about 2 years ago - 3 comments

#282 - question about expected performance of quantized operations

Issue - State: open - Opened by carsonswope about 2 years ago - 4 comments

#282 - question about expected performance of quantized operations

Issue - State: open - Opened by carsonswope about 2 years ago - 3 comments

#262 - How to measure the compilation time of dml graph on GPU?

Issue - State: open - Opened by mingmingtasd over 2 years ago - 5 comments

#256 - Bump numpy from 1.18.1 to 1.22.0 in /TensorFlow/TF1/squeezenet/src

Pull Request - State: open - Opened by dependabot[bot] over 2 years ago
Labels: dependencies

#247 - PyTorch/TensorFlow C++ APIs?

Issue - State: open - Opened by DrewRidley over 2 years ago - 6 comments
Labels: pytorch-directml, tensorflow

#224 - [SUPPORT REQUEST] 3D Convolution

Issue - State: open - Opened by hyperfraise over 2 years ago - 5 comments
Labels: pytorch-directml

#213 - Bump pillow from 7.0.0 to 9.0.1 in /TensorFlow/squeezenet/src

Pull Request - State: open - Opened by dependabot[bot] over 2 years ago - 16 comments
Labels: dependencies

#212 - GEMM operator with variable size input

Issue - State: open - Opened by jacobgorm over 2 years ago - 19 comments

#211 - Pytorch DirectML integration in Pytorch Lightning

Issue - State: open - Opened by sto1 over 2 years ago - 5 comments
Labels: pytorch-directml

#202 - Bump protobuf from 3.11.3 to 3.15.0 in /TensorFlow/squeezenet/src

Pull Request - State: open - Opened by dependabot[bot] almost 3 years ago - 16 comments
Labels: dependencies

#196 - RuntimeError: don't know how to restore data location of torch.FloatStorage

Issue - State: open - Opened by masc-it almost 3 years ago - 14 comments
Labels: pytorch-directml

#177 - .to('dml') not working

Issue - State: closed - Opened by mpenn114 almost 3 years ago - 7 comments
Labels: pytorch-directml

#177 - .to('dml') not working

Issue - State: closed - Opened by mpenn114 almost 3 years ago - 7 comments
Labels: pytorch-directml

#177 - .to('dml') not working

Issue - State: closed - Opened by mpenn114 almost 3 years ago - 7 comments
Labels: pytorch-directml

#172 - No matching distribution found for pytorch-directml

Issue - State: closed - Opened by oyyleung about 3 years ago - 5 comments
Labels: pytorch-directml

#157 - Bump opencv-python from 4.1.1.26 to 4.2.0.32 in /TensorFlow/yolov3

Pull Request - State: open - Opened by dependabot[bot] about 3 years ago - 16 comments
Labels: dependencies

#131 - Bump urllib3 from 1.25.8 to 1.26.5 in /TensorFlow/squeezenet/src

Pull Request - State: open - Opened by dependabot[bot] over 3 years ago - 16 comments
Labels: dependencies

#129 - unable to build pydirectml

Issue - State: closed - Opened by iperov over 3 years ago - 1 comment

#128 - can you please publish python wheels of pydirectml bindings?

Issue - State: closed - Opened by iperov over 3 years ago

#127 - prebuilt onnxruntime with direct ml?

Issue - State: closed - Opened by iperov over 3 years ago - 10 comments

#108 - DirectML is x2.8 slower than CUDA

Issue - State: closed - Opened by iperov over 3 years ago - 38 comments

#107 - Notification: "NVD3D10: CPU cyclestats are disabled on client virtualization"

Issue - State: closed - Opened by markwny over 3 years ago - 3 comments

#100 - Using DirectML from C#

Issue - State: open - Opened by vishvanathan-k over 3 years ago - 5 comments

#100 - Using DirectML from C#

Issue - State: open - Opened by vishvanathan-k over 3 years ago - 5 comments

#63 - DirectML v1.4 ARM64 version?

Issue - State: open - Opened by zorderjohn almost 4 years ago - 4 comments

GitHub / microsoft/DirectML issues and pull requests