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
libc++abi: terminating with uncaught exception of type c10::Error: allocated_ <= ARENA_MAX_SIZE INTERNAL ASSERT FAILED at "/stuff/pytorch/pytorch/pytorch/functorch/csrc/dim/arena.h":227, please report a bug to PyTorch.
The text was updated successfully, but these errors were encountered:
I don't understand this yet, but can confirm that this appears to be unrelated to tensordict.
importtorcht=torch.rand(6401, 1)
# build splitstotal=0l= []
whiletotal<6400:
l.append(torch.randint(2, 10, (1,)).item())
total+=l[-1]
# could get RuntimeError if `sum(l) != 6401`l[-1] =6401-sum(l[:-1])
# this is fine_=t.split(l, 0)
But now, if we import functorch.dim it will fail
importfunctorch.dim_=t.split(l, 0)
So something is happening at import with functorch that results in this bad behaviour. I'll try to pin it down a bit more but seems clear it's an issue upstream.
vmoens
changed the title
[BUG]
[BUG] functorch.dim breaks split of big tensors
Nov 7, 2022
Describe the bug
For some reason (probably unrealted to tensordict), functorch dims fails with big tensors.
This should result in an error like this:
The text was updated successfully, but these errors were encountered: