Lightning-AI / lightning-thunder

Make PyTorch models up to 40% faster! Thunder is a source to source compiler for PyTorch. It enables using different hardware executors at once; across one or thousands of GPUs.
Apache License 2.0
1.2k stars 80 forks source link

`torch.exp_` is not suppurted, but `torch.Tensor.exp_` is #1072

Closed crcrpar closed 2 months ago

crcrpar commented 2 months ago

Note: If you have a model or program that is not supported yet but should be, please use the program coverage template.

🐛 Bug

To Reproduce

import torch

from thunder.examine import examine

def f(a):
    return torch.exp_(a)

def g(a):
    return a.exp_()

a = torch.randn(2)
examine(f, a)
examine(g, a)

then python a.py gives

Found 1 distinct operations, of which 0 (0.0%) are supported
Please file an issue requesting the following operators here: https://github.com/Lightning-AI/lightning-thunder/issues/new
_VariableFunctionsClass.exp_ of torch
Found 1 distinct operations, of which 1 (100.0%) are supported
The function appears to be working as expected