feat: 支持百度千帆向量模型

apps/setting/models_provider/impl/wenxin_model_provider/credential/embedding.py

@@ -0,0 +1,42 @@
+# coding=utf-8
+"""
+    @project: MaxKB
+    @Author：虎
+    @file： embedding.py
+    @date：2024/10/17 15:40
+    @desc:
+"""
+from typing import Dict
+
+from common import forms
+from common.exception.app_exception import AppApiException
+from common.forms import BaseForm
+from setting.models_provider.base_model_provider import BaseModelCredential, ValidCode
+
+
+class QianfanEmbeddingCredential(BaseForm, BaseModelCredential):
+
+    def is_valid(self, model_type: str, model_name, model_credential: Dict[str, object], provider,
+                 raise_exception=False):
+        model_type_list = provider.get_model_type_list()
+        if not any(list(filter(lambda mt: mt.get('value') == model_type, model_type_list))):
+            raise AppApiException(ValidCode.valid_error.value, f'{model_type} 模型类型不支持')
+        self.valid_form(model_credential)
+        try:
+            model = provider.get_model(model_type, model_name, model_credential)
+            model.embed_query('你好')
+        except Exception as e:
+            if isinstance(e, AppApiException):
+                raise e
+            if raise_exception:
+                raise AppApiException(ValidCode.valid_error.value, f'校验失败,请检查参数是否正确: {str(e)}')
+            else:
+                return False
+        return True
+
+    def encryption_dict(self, model: Dict[str, object]):
+        return {**model, 'qianfan_sk': super().encryption(model.get('qianfan_sk', ''))}
+
+    qianfan_ak = forms.PasswordInputField('API Key', required=True)
+
+    qianfan_sk = forms.PasswordInputField("Secret Key", required=True)

apps/setting/models_provider/impl/wenxin_model_provider/model/embedding.py

@@ -0,0 +1,23 @@
+# coding=utf-8
+"""
+    @project: MaxKB
+    @Author：虎
+    @file： embedding.py
+    @date：2024/10/17 16:48
+    @desc:
+"""
+from typing import Dict
+
+from langchain_community.embeddings import QianfanEmbeddingsEndpoint
+
+from setting.models_provider.base_model_provider import MaxKBBaseModel
+
+
+class QianfanEmbeddings(MaxKBBaseModel, QianfanEmbeddingsEndpoint):
+    @staticmethod
+    def new_instance(model_type, model_name, model_credential: Dict[str, object], **model_kwargs):
+        return QianfanEmbeddings(
+            model=model_name,
+            qianfan_ak=model_credential.get('qianfan_ak'),
+            qianfan_sk=model_credential.get('qianfan_sk'),
+        )

apps/setting/models_provider/impl/wenxin_model_provider/wenxin_model_provider.py

@@ -11,11 +11,14 @@
 from common.util.file_util import get_file_content
 from setting.models_provider.base_model_provider import ModelProvideInfo, ModelTypeConst, ModelInfo, IModelProvider, \
     ModelInfoManage
+from setting.models_provider.impl.wenxin_model_provider.credential.embedding import QianfanEmbeddingCredential
 from setting.models_provider.impl.wenxin_model_provider.credential.llm import WenxinLLMModelCredential
+from setting.models_provider.impl.wenxin_model_provider.model.embedding import QianfanEmbeddings
 from setting.models_provider.impl.wenxin_model_provider.model.llm import QianfanChatModel
 from smartdoc.conf import PROJECT_DIR
 
 win_xin_llm_model_credential = WenxinLLMModelCredential()
+qianfan_embedding_credential = QianfanEmbeddingCredential()
 model_info_list = [ModelInfo('ERNIE-Bot-4',
                              'ERNIE-Bot-4是百度自行研发的大语言模型，覆盖海量中文数据，具有更强的对话问答、内容创作生成等能力。',
                              ModelTypeConst.LLM, win_xin_llm_model_credential, QianfanChatModel),
@@ -41,13 +44,16 @@
                              '千帆团队在Llama-2-7b基础上的中文增强版本，在CMMLU、C-EVAL等中文知识库上表现优异。',
                              ModelTypeConst.LLM, win_xin_llm_model_credential, QianfanChatModel)
                    ]
-
+embedding_model_info = ModelInfo('Embedding-V1',
+                                 'Embedding-V1是一个基于百度文心大模型技术的文本表示模型，可以将文本转化为用数值表示的向量形式，用于文本检索、信息推荐、知识挖掘等场景。 Embedding-V1提供了Embeddings接口，可以根据输入内容生成对应的向量表示。您可以通过调用该接口，将文本输入到模型中，获取到对应的向量表示，从而进行后续的文本处理和分析。',
+                                 ModelTypeConst.EMBEDDING, qianfan_embedding_credential, QianfanEmbeddings)
 model_info_manage = ModelInfoManage.builder().append_model_info_list(model_info_list).append_default_model_info(
     ModelInfo('ERNIE-Bot-4',
               'ERNIE-Bot-4是百度自行研发的大语言模型，覆盖海量中文数据，具有更强的对话问答、内容创作生成等能力。',
               ModelTypeConst.LLM,
               win_xin_llm_model_credential,
-              QianfanChatModel)).build()
+              QianfanChatModel)).append_model_info(embedding_model_info).append_default_model_info(
+    embedding_model_info).build()
 
 
 class WenxinModelProvider(IModelProvider):