Yukio Siraichi
Yukio Siraichi
No. It works as expected by replacing `torch_xla.amp` by `torch.amp`.
Printing multiplication by negative number with custom (infix) function not correctly parenthesized.
Hmm. I guess that could work. The problem I see, however, is the way `_print_Mul` gets its precedence. https://github.com/sympy/sympy/blob/9817721589a769319c22437c116d2deffce26aef/sympy/printing/str.py#L263 This function returns the `Add` precedence due to the leading negative...
Weekly update (Dec 1~Dec 10): ### Models fixed: - DALLE2_pytorch - training is now failing with the same error as inductor - stable_diffusion_unet - training is still failing with OOM...
Weekly update (Dec 11~Dec 15): ### Models fixed: - pytorch_CycleGAN_and_pix2pix - nvidia_deeprecommender - dynamo+openxla training is still failling - simple_gpt and simple_gpt_tp_manual - failing due to the same reasons as...
Weekly update (Jan 8 ~ Jan 12): ### Pass rate (out of 99 benchmarks): | | Inference | Training | |----|----------------|--------------| | Inductor | 91 | 64 | | Non-Dynamo...
Weekly update (Jan 15 ~ Jan 19): ### Pass rate (out of 99 benchmarks): | | Inference | Training | |----|----------------|--------------| | Inductor | 85 | 62 | | Non-Dynamo...
Weekly update (Jan 22 ~ Jan 26): ### Pass rate (out of 99 benchmarks): | | Inference | Training | |----|----------------|--------------| | Inductor | 88 | 63 | | Non-Dynamo...
Weekly update (Jan 29 ~ Feb 2): ### Pass rate (out of 99 benchmarks): #### A100 | | Inference | Training | |----|----------------|--------------| | Inductor | 87 (last: 88) |...
Weekly update (Feb 5 ~ Feb 9): ### Pass rate (out of 99 benchmarks): #### A100 | | Inference | Training | |----|----------------|--------------| | Inductor | 87 (last: 87) |...
Weekly update (Feb 12 ~ Feb 16): ### Pass rate (out of 99 benchmarks): Could not run the benchmarks this time, due to a compilation issue: #6564 ----- ### PRs...