Uses torch._inductor.compile instead of torch.compile in benchmark script to avoid segmentation by Dynamo #1540
Azure Pipelines / lightning-thunder (ipynb) (jupyter ubuntu22.04 | cuda 12.1 | torch-nightly)
succeeded
Dec 11, 2024 in 7m 38s
jupyter ubuntu22.04 | cuda 12.1 | torch-nightly succeeded
Loading