Hi, I am using Lightning CLI to configure my experiments and i want to know the recommended practice to use torch.compile with Lightning CLI. By the way, I am 16-mixed in Lightning with RTX3090, and i got the following warning
You are using a CUDA device ('NVIDIA GeForce RTX 3090') that has Tensor Cores. To properly utilize them, you should set `torch.set_float32_matmul_precision('medium' | 'high')` which will trade-off precision for performance. For more details, read https://pytorch.org/docs/stable/generated/torch.set_float32_matmul_precision.html#torch.set_float32_matmul_precision
what should i do in Lighting ? Thanks so much.