You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Turns out nvfuser's randlike function generates "1.0" in uniform samples while in PyTorch uniform distribution is sampled on the interval [0.0, 1.0), "0.0" is included, "1.0" is excluded.
I think nvfuser's random numbers should be changed to sample on [0.0, 1.0) interval. uniform and uniformf functions are implemented here, and used for randLike device functions: https://github.com/csarofeen/pytorch/blob/fa4e6a4739a9daaa0e4111fb4730704d79c91010/torch/csrc/jit/codegen/cuda/runtime/helpers.cu#L302-L308
We could replace generated samples of 1.0 with 0.0:
🐛 Describe the bug
See the discussion here #1799 (comment).
Turns out nvfuser's randlike function generates "1.0" in uniform samples while in PyTorch uniform distribution is sampled on the interval [0.0, 1.0), "0.0" is included, "1.0" is excluded.
I think nvfuser's random numbers should be changed to sample on [0.0, 1.0) interval.
uniform
anduniformf
functions are implemented here, and used forrandLike
device functions: https://github.com/csarofeen/pytorch/blob/fa4e6a4739a9daaa0e4111fb4730704d79c91010/torch/csrc/jit/codegen/cuda/runtime/helpers.cu#L302-L308We could replace generated samples of 1.0 with 0.0:
Versions
devel branch
The text was updated successfully, but these errors were encountered: