Skip to content

Push Message formatting into _gen_model_input #1295

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
merged 2 commits into from
Oct 14, 2024
Merged

Push Message formatting into _gen_model_input #1295

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
482b311
Revert Generate Behavior for non-Flamingo Models
Jack-Khuu Oct 11, 2024
a505a59
Simplify
Jack-Khuu Oct 11, 2024
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
7 changes: 6 additions & 1 deletion torchchat/generate.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -797,6 +797,10 @@ def _gen_model_input(
max_new_tokens is not None
), "max_new_tokens must be specified for Flamingo models"

# Wrap string prompts into a list
if isinstance(prompt, str):
prompt = [{"role": "user", "content": prompt}]

image_found = False
messages = []
for message in prompt:
Expand Down Expand Up @@ -959,8 +963,9 @@ def chat(
max_seq_length = (
text_transformer_args.max_seq_length if text_transformer_args else 2048
)

encoded, batch = self._gen_model_input(
[{"role": "user", "content": generator_args.prompt}],
generator_args.prompt,
generator_args.image_prompts,
generator_args.max_new_tokens,
max_seq_length,
Expand Down
Loading