Skip to content

Fix missing valid_sequence_lengths tensor from generation_config #322

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
wszczurekhabana merged 1 commit into from
Jul 31, 2024
Merged

Fix missing valid_sequence_lengths tensor from generation_config #322

Show file tree
Hide file tree
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
2 changes: 1 addition & 1 deletion examples/text-generation/run_generation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -610,7 +610,7 @@ def compute_valid_sequence_lengths_tensor(input_tokens):
attn_mask = input_tokens['attention_mask']
return torch.sum(attn_mask, dim=1)
valid_sequence_lengths = compute_valid_sequence_lengths_tensor(input_tokens).to(args.device)
setattr(generation_config, 'valid_sequence_lengths', valid_sequence_lengths)
generation_config.valid_sequence_lengths = valid_sequence_lengths
else:
input_tokens = tokenizer.batch_encode_plus(input_sentences, return_tensors="pt", padding=True)
encode_duration = time.perf_counter() - encode_t0
Expand Down
1 change: 1 addition & 0 deletions examples/text-generation/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -601,6 +601,7 @@ def setup_generation_config(args, model, assistant_model, tokenizer):
generation_config.flash_attention_causal_mask = args.flash_attention_causal_mask
generation_config.flash_attention_fast_softmax = args.flash_attention_fast_softmax
generation_config.trust_remote_code = args.trust_remote_code
setattr(generation_config, 'valid_sequence_lengths', None)

return generation_config

Expand Down