From ca7375ac20fc831516e77a24f6d782bb6e22b021 Mon Sep 17 00:00:00 2001 From: ZhangShenao <15201440436@163.com> Date: Wed, 13 Nov 2024 02:57:35 +0800 Subject: [PATCH] Improvement[Community]Improve Embeddings API (#28038) - Fix `BaichuanTextEmbeddings` api url - Remove unused params in api doc - Fix word spelling --- libs/community/langchain_community/embeddings/baichuan.py | 4 ++-- libs/community/langchain_community/embeddings/dashscope.py | 2 -- libs/community/langchain_community/embeddings/fake.py | 3 ++- 3 files changed, 4 insertions(+), 5 deletions(-) diff --git a/libs/community/langchain_community/embeddings/baichuan.py b/libs/community/langchain_community/embeddings/baichuan.py index a2f63773a27..c12aaa44f1c 100644 --- a/libs/community/langchain_community/embeddings/baichuan.py +++ b/libs/community/langchain_community/embeddings/baichuan.py @@ -15,7 +15,7 @@ from pydantic import ( from requests import RequestException from typing_extensions import Self -BAICHUAN_API_URL: str = "http://api.baichuan-ai.com/v1/embeddings" +BAICHUAN_API_URL: str = "https://api.baichuan-ai.com/v1/embeddings" # BaichuanTextEmbeddings is an embedding model provided by Baichuan Inc. (https://www.baichuan-ai.com/home). # As of today (Jan 25th, 2024) BaichuanTextEmbeddings ranks #1 in C-MTEB @@ -25,7 +25,7 @@ BAICHUAN_API_URL: str = "http://api.baichuan-ai.com/v1/embeddings" # Official Website: https://platform.baichuan-ai.com/docs/text-Embedding # An API-key is required to use this embedding model. You can get one by registering # at https://platform.baichuan-ai.com/docs/text-Embedding. -# BaichuanTextEmbeddings support 512 token window and preduces vectors with +# BaichuanTextEmbeddings support 512 token window and produces vectors with # 1024 dimensions. diff --git a/libs/community/langchain_community/embeddings/dashscope.py b/libs/community/langchain_community/embeddings/dashscope.py index 3268e5de3bd..59bd76e0de8 100644 --- a/libs/community/langchain_community/embeddings/dashscope.py +++ b/libs/community/langchain_community/embeddings/dashscope.py @@ -143,8 +143,6 @@ class DashScopeEmbeddings(BaseModel, Embeddings): Args: texts: The list of texts to embed. - chunk_size: The chunk size of embeddings. If None, will use the chunk size - specified by the class. Returns: List of embeddings, one for each text. diff --git a/libs/community/langchain_community/embeddings/fake.py b/libs/community/langchain_community/embeddings/fake.py index 67dc1d6b6f5..6bbfeeb45cd 100644 --- a/libs/community/langchain_community/embeddings/fake.py +++ b/libs/community/langchain_community/embeddings/fake.py @@ -36,7 +36,8 @@ class DeterministicFakeEmbedding(Embeddings, BaseModel): np.random.seed(seed) return list(np.random.normal(size=self.size)) - def _get_seed(self, text: str) -> int: + @staticmethod + def _get_seed(text: str) -> int: """ Get a seed for the random generator, using the hash of the text. """