Mikayla Gawarecki
PyTorch module maintainer for torch.nn and the C++ API; author of PyTorch tutorials
- Module maintainer, NN APIs (torch.nn) and C++ API (libtorch + StableABI), PyTorch (open-source project)1
Mikayla Gawarecki is listed on PyTorch's maintainers page as a module maintainer for the NN APIs (torch.nn) and the C++ API (libtorch and StableABI), and as an emeritus maintainer of NestedTensor (torch.nested). Gawarecki wrote the PyTorch tutorial "Accelerating PyTorch Transformers by replacing nn.Transformer with Nested Tensors and torch.compile()," which covers nested tensors, scaled_dot_product_attention, torch.compile, and FlexAttention, and the recipe "Extension points in nn.Module for load_state_dict and tensor subclasses" (April 2024).123
Work
PyTorch (torch.nn and the C++ API)Open-source code
Listed as a maintainer of torch.nn and of the C++ API (libtorch + StableABI) and an emeritus maintainer of torch.nested. The repository is under a BSD-style license.14
PyTorch tutorials on nested tensors and nn.Module extension pointsPublished research
Credited author of the tutorial "Accelerating PyTorch Transformers by replacing nn.Transformer with Nested Tensors and torch.compile()" and of the recipe "Extension points in nn.Module for load_state_dict and tensor subclasses," which introduces torch.utils.swap_tensors.23
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.