Skip to content
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

Jump to bottom

fix on blip_caption.py: prepare prompt for beam search, aovid tensor … #251

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix on blip_caption.py: prepare prompt for beam search, aovid tensor … #251

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.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
4 changes: 4 additions & 0 deletions lavis/models/blip_models/blip_caption.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -183,6 +183,10 @@ def generate(
prompt = self.tokenizer(prompt, return_tensors="pt").to(self.device)
prompt.input_ids[:, 0] = self.tokenizer.bos_token_id
prompt.input_ids = prompt.input_ids[:, :-1]

# prepare prompt for beam search
if not use_nucleus_sampling:
prompt.input_ids = torch.repeat_interleave(prompt.input_ids, num_beams, dim=0)

# get decoded text
decoder_out = self.text_decoder.generate_from_encoder(
Expand Down