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.

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

[python] Makes gpt-j model use triton mode #827

Merged
merged 1 commit into from
Jun 12, 2023

Conversation

frankfliu
Copy link
Contributor

Description

Brief description of what this PR is about

  • If this change is a backward incompatible change, why must this change be made?
  • Interesting edge cases to note here

@frankfliu frankfliu requested review from zachgk and a team as code owners June 10, 2023 23:24
output_len = parameters.pop("max_seq_len", max_length)
if self.use_triton:
output_length = [output_len] * len(input_data)
result = self.model.pipeline_generate(input_data,
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

append with **parameters

@frankfliu frankfliu changed the title [python] Makes triton model default to FasterTransformer [python] Makes gpt-j model use triton mode Jun 12, 2023
"xlm-roberta");

private static final List<String> FASTERTRANSFORMER_MODELS =
List.of("bloom", "gpt2", "t5", "opt");
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

add gpt_neox

@frankfliu frankfliu merged commit ca5768f into deepjavalibrary:master Jun 12, 2023
8 checks passed
@frankfliu frankfliu deleted the ft branch June 12, 2023 19:31
KexinFeng pushed a commit to KexinFeng/djl-serving-forked that referenced this pull request Aug 16, 2023
KexinFeng pushed a commit to KexinFeng/djl-serving-forked that referenced this pull request Aug 16, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants