Skip to content

Support max_seq_len argument in multimodal text-to-text export #209

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

Open
kamalkraj wants to merge 1 commit into
base: main
Choose a base branch
from
+2 −1
Open

Support max_seq_len argument in multimodal text-to-text export #209

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
3 changes: 2 additions & 1 deletion optimum/exporters/executorch/tasks/multimodal_text_to_text.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,7 +62,8 @@ def load_multimodal_text_to_text_model(model_name_or_path: str, **kwargs):
attn_implementation = kwargs.get("attn_implementation", "custom_sdpa" if use_custom_sdpa else "sdpa")
cache_implementation = kwargs.get("cache_implementation", "static")
use_custom_sdpa = use_custom_sdpa or attn_implementation == "custom_sdpa"
max_length = kwargs.get("max_length", 2048)
max_seq_len = kwargs.get("max_seq_len", None)
max_length = max_seq_len if max_seq_len is not None else kwargs.get("max_length", 2048)
config = kwargs.get("config") or AutoConfig.from_pretrained(model_name_or_path)

# Load preprocessor_config.json if it exists
Expand Down