-
Notifications
You must be signed in to change notification settings - Fork 439
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
make_simple_dynamo_backend
does not work for single argument models.
#3204
Comments
I'm not aware of anyone who is actively working on/maintaining these paths. Started a thread on discord as the conversation about what to mark as deprecated has looped a couple of times: https://discord.com/channels/636084430946959380/742573221882364009/1232091021450547221 |
(but if you see a fix, patches welcome) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Description
make_simple_dynamo_backend
does not work for single argument models. To support multiple args, it takes in*inputs
. In case of a singletorch.Tensor
input it will unpack it instead of handling it in its entirety.Specifically for the example below you will get the following error:
Example
For the sake of brevity, this example returns
gm.forward
and does not usetorch-mlir
in the backend implementation. However, the bug is also present whentorch-mlir
is used in the backend.Possible solution
In the file
torch_mlir/dynamo.py
thedynamo_callable
function takes in*inputs
. This should only be the case iflen(example_inputs) > 1
.The text was updated successfully, but these errors were encountered: