Skip to content

Check inference input_id tokens length #4349

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.

Sign up for GitHub

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

Merged
awan-10 merged 1 commit into from
Sep 19, 2023
Merged

Check inference input_id tokens length #4349

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions deepspeed/inference/engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -598,4 +598,12 @@ def _generate(self, *inputs, **kwargs):
raise NotImplementedError("DeepSpeed does not support `num_beams` > 1, if this is important to you please "
"add your request to: https://github.com/microsoft/DeepSpeed/issues/2506")

if ("input_ids" in kwargs) and (kwargs["input_ids"].dim() == 2):
for input_tensor in kwargs["input_ids"]:
tensor_length = input_tensor.shape[-1]
if tensor_length > self._config.max_out_tokens:
raise RuntimeError(
f"Input with size {tensor_length} exceeds maximum length of {self._config.max_out_tokens}. Please increase `max_tokens` in the DeepSpeed Inference Config."
)

return self.module.generate(*inputs, **kwargs)