openvino_pytorch_layers
openvino_pytorch_layers copied to clipboard
How to export PyTorch models with unsupported layers to ONNX and then to Intel OpenVINO
Hi @dkurt , Your solution to overcome a pytorch to IR model conversion issue [here](https://github.com/openvinotoolkit/training_extensions/pull/630#issuecomment-1023967160) was really helpful, however we couldn't proceed with the PR due to dataset license issues....
[torch.lstsq](https://pytorch.org/docs/stable/generated/torch.lstsq.html) accepts `B` and `A` matrices of shapes `MxK` and `MxN` correspondingly. This PR enables least-squares solution for https://github.com/EIDOSlab/torchstain/blob/main/torchstain/normalizers/torch_macenko_normalizer.py where `K` is always equals `2`.
Hi, @dkurt Thanks for your sharing! I wonder whether this op works for other arguments? For example, F.grid_sample(img, grid, mode='bilinear', padding_mode='reflection', align_corners=True). Will it be supported in iGPU in the...