From b4ef50bdb5c54f7aded42bedf79c0a3ad471bbcf Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E9=BB=84=E8=85=BE?= <101850389+hangters@users.noreply.github.com> Date: Fri, 16 Aug 2024 09:49:27 +0800 Subject: [PATCH] fix OpenAI Embedding length error (#1972) ### What problem does this PR solve? #1958 fix OpenAI Embedding length error ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue) --------- Co-authored-by: Zhedong Cen --- rag/llm/embedding_model.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/rag/llm/embedding_model.py b/rag/llm/embedding_model.py index d0e011197..cda6b2429 100644 --- a/rag/llm/embedding_model.py +++ b/rag/llm/embedding_model.py @@ -99,14 +99,14 @@ class OpenAIEmbed(Base): self.model_name = model_name def encode(self, texts: list, batch_size=32): - texts = [truncate(t, 8196) for t in texts] + texts = [truncate(t, 8191) for t in texts] res = self.client.embeddings.create(input=texts, model=self.model_name) return np.array([d.embedding for d in res.data] ), res.usage.total_tokens def encode_queries(self, text): - res = self.client.embeddings.create(input=[truncate(text, 8196)], + res = self.client.embeddings.create(input=[truncate(text, 8191)], model=self.model_name) return np.array(res.data[0].embedding), res.usage.total_tokens