Accelerating Complex-Valued Tensors With Torch.compile - Hameer Abbasi, OpenTeams Inc.
About this talk
This talk covers the advancements in supporting complex-valued tensors with Torch.compile, presented by Hameer Abbasi from OpenTeams Inc. The speaker discusses the importance of Torch.compile in enhancing performance across various machine learning and scientific computing applications while addressing its limitations, particularly the lack of support for complex-valued tensors. He explores the challenges faced in integrating this functionality and outlines goals for future improvements, including potential benefits like reducing binary size through just-in-time compilation of kernels.
More from this event
See all 103 talks →
What PyTorch Conference Europe 2026 Was Really Like – Official PyTorchCon EU Highlights | Paris
0:53
Lightning Talk: How DeepInverse Is Solving Imaging in Science and H... Andrew Wang & Minh Hai Nguyen
9:50
Why WideEP Inference Needs Data-Parallel-Aware Scheduling - Maroon Ayoub & Tyler Michael Smith
25:37
Write Once, Run Everywhere with Pytorch Transformers - Pedro Cuenca, Hugging Face
19:17