bisheng-langchain 0.3.0rc0__py3-none-any.whl → 0.3.1__py3-none-any.whl

bisheng_langchain/rag/scoring/ragas_score.py

@@ -0,0 +1,183 @@
+import json
+import os
+import re
+from collections import defaultdict
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Tuple, Union
+
+import pandas as pd
+from datasets import Dataset
+from loguru import logger
+from ragas import evaluate
+from ragas.metrics import AnswerCorrectness, AnswerCorrectnessBisheng, AnswerRecallBisheng
+
+
+@dataclass
+class RagScore:
+    excel_path: str
+    save_path: str
+    question_column: str
+    gt_column: str
+    answer_column: str
+    metrics: List[str]
+    contexts_column: Optional[str] = None
+    query_type_column: Optional[str] = None
+    gt_split_column: Optional[str] = None
+    batch_size: int = 5
+    whether_gtsplit: bool = False
+
+    def _validate_metrics(self):
+        for metric in self.metrics:
+            if not hasattr(self, f'ragas_{metric}'):
+                raise Exception(f'"ragas_{metric}" 未实现!')
+
+    def ragas_answer_correctness(self, dataset: Dataset) -> pd.DataFrame:
+        # answer_correctness, 只考虑事实相似度
+        weights = [1.0, 0.0]
+        answer_correctness = AnswerCorrectness(weights=weights, batch_size=self.batch_size)
+        result = evaluate(
+            dataset=dataset,
+            metrics=[
+                answer_correctness,
+            ],
+        )
+        self.score_map_keys = list(result.keys())
+        df = result.to_pandas()
+        return df
+
+    def ragas_answer_correctness_bisheng(self, dataset: Dataset) -> pd.DataFrame:
+        answer_correctness = AnswerCorrectnessBisheng(batch_size=self.batch_size)
+        result = evaluate(
+            dataset=dataset,
+            metrics=[answer_correctness],
+        )
+        self.score_map_keys = list(result.keys())
+        df = result.to_pandas()
+        return df
+
+    def ragas_answer_recall_bisheng(self, dataset: Dataset) -> pd.DataFrame:
+        answer_recall =AnswerRecallBisheng(batch_size=self.batch_size, 
+                                           whether_gtsplit=self.whether_gtsplit)
+        result = evaluate(
+            dataset=dataset,
+            metrics=[answer_recall],
+        )
+        self.score_map_keys = list(result.keys())
+        df = result.to_pandas()
+        return df
+
+    def _remove_source(self, pred: str) -> str:
+        """去除【1†source】, only for openai assistant"""
+        pattern = re.compile("【(\d+)†source】")
+        match = re.findall(pattern, pred)
+        for i in match:
+            str_temp = f"【{i}†source】"
+            pred = pred.replace(str_temp, '')
+        return pred
+
+    def score(self) -> Tuple[pd.DataFrame, Optional[pd.DataFrame]]:
+        df = pd.read_excel(self.excel_path)
+        ori_row_nums = df.shape[0]
+
+        # 删除含有na的行
+        columns_to_check = [
+            self.question_column,
+            self.gt_column,
+            self.answer_column,
+            self.contexts_column,
+            self.query_type_column,
+        ]
+        # 是否有要点拆分列
+        if self.gt_split_column:
+            columns_to_check.append(self.gt_split_column)
+
+        df.dropna(subset=[col for col in columns_to_check if col], inplace=True)
+        df = df.reset_index()
+        print(f'删除含有na的行 {ori_row_nums - df.shape[0]} 个!')
+        print(f'总计 {df.shape[0]} 个问题')
+
+        questions = df[self.question_column].tolist()
+        answers = df[self.answer_column].tolist()
+        # answers = df[self.answer_column].apply(self._remove_source).tolist() # for openai assistant
+        ground_truths = df[self.gt_column].apply(lambda x: [x]).tolist()
+        # todo: contexts可能是保存在json中的，这段代码可能需要修改
+        contexts = (
+            [['']] * len(questions)
+            if not self.contexts_column
+            else df[self.contexts_column].apply(lambda x: [x]).tolist()
+        )
+        # To dict
+        if self.gt_split_column:
+            gtsplit = df[self.gt_split_column].tolist()
+            data: Dict[str, List[Any]] = {
+                "question": questions,
+                "answer": answers,
+                "contexts": contexts,
+                "ground_truths": ground_truths,
+                'gt_split_point': gtsplit
+            }
+        else:
+            data: Dict[str, List[Any]] = {
+                "question": questions,
+                "answer": answers,
+                "contexts": contexts,
+                "ground_truths": ground_truths,
+            }
+        # Convert dict to dataset
+        dataset = Dataset.from_dict(data)
+
+        self._validate_metrics()
+
+        save_group_df = dict()
+        for metric_name in self.metrics:
+            ragas_result = getattr(self, f'ragas_{metric_name}')(dataset)
+            if metric_name =='answer_recall_bisheng':
+                if self.gt_split_column:
+                    df[self.gt_split_column] = ragas_result["gt_split_point"]
+                else:
+                    df["gt_split_point"] = ragas_result["gt_split_point"]
+                df["analyse"] = ragas_result["analyse"]
+
+            score_map = dict().fromkeys(self.score_map_keys, ragas_result)
+            for metric, scores in score_map.items():
+                df[metric] = df.index.map({idx: rows[metric] for idx, rows in scores.iterrows()})
+
+            if self.query_type_column and self.query_type_column in df.columns:
+                grouped_df = df.groupby(self.query_type_column)
+                grouped_df = grouped_df.agg({self.question_column: 'count', **{metric: 'mean' for metric in score_map}})
+                grouped_df.rename(columns={self.question_column: '问题个数'}, inplace=True)
+                
+                total_question = grouped_df['问题个数'].sum()
+                grouped_df.loc['总计', '问题个数'] = total_question
+                for metric in score_map:
+                    grouped_df.loc['总计', metric] = df[metric].sum() / total_question
+                save_group_df[f'{metric_name}_group'] = grouped_df
+                
+                print(grouped_df.to_markdown())
+
+        # save
+        output_path = Path(self.save_path) / f"{Path(self.excel_path).stem}_score.xlsx"
+        with pd.ExcelWriter(output_path) as writer:
+            df.to_excel(writer, sheet_name='Sheet1', index=False)
+            if len(save_group_df):
+                for metric, grouped_df in save_group_df.items():
+                    grouped_df.to_excel(writer, sheet_name=metric, index=True)
+        print(f'保存到 {output_path} 成功!')
+
+
+if __name__ == '__main__':
+    params = {
+        'excel_path': '/home/gulixin/workspace/llm/bisheng/src/bisheng-langchain/experimental/rag/data/test.xlsx',
+        'save_path': '/home/gulixin/workspace/llm/bisheng/src/bisheng-langchain/experimental/rag/data',
+        'question_column': '问题',
+        'gt_column': 'GT',
+        'answer_column': 'rag_answer',
+        'query_type_column': '问题类型',
+        # 'metrics': ['answer_correctness_bisheng'],
+        'metrics': ['answer_recall_bisheng'],
+        'batch_size': 10,
+        'whether_gtsplit': False,
+    }
+    rag_score = RagScore(**params)
+    rag_score.score()

bisheng_langchain/rag/utils.py

@@ -0,0 +1,181 @@
+# This module is used to import any langchain class by name.
+
+import importlib
+from typing import Any, Type
+
+from langchain.agents import Agent
+from langchain.base_language import BaseLanguageModel
+from langchain.chains.base import Chain
+from langchain.chat_models.base import BaseChatModel
+from langchain.prompts import PromptTemplate
+from langchain.tools import BaseTool
+
+
+def import_module(module_path: str) -> Any:
+    """Import module from module path"""
+    if 'from' not in module_path:
+        # Import the module using the module path
+        return importlib.import_module(module_path)
+    # Split the module path into its components
+    _, module_path, _, object_name = module_path.split()
+
+    # Import the module using the module path
+    module = importlib.import_module(module_path)
+
+    return getattr(module, object_name)
+
+
+def import_class(class_path: str) -> Any:
+    """Import class from class path"""
+    module_path, class_name = class_path.rsplit('.', 1)
+    module = import_module(module_path)
+    return getattr(module, class_name)
+
+
+def import_by_type(_type: str, name: str) -> Any:
+    from bisheng_langchain import chat_models
+    """Import class by type and name"""
+    if _type is None:
+        raise ValueError(f'Type cannot be None. Check if {name} is in the config file.')
+    func_dict = {
+        'agents': import_agent,
+        'prompts': import_prompt,
+        'llms': {
+            'llm': import_llm,
+            'chat': import_chat_llm,
+            'contribute': import_chain_contribute_llm
+        },
+        'tools': import_tool,
+        'chains': import_chain,
+        'toolkits': import_toolkit,
+        'memory': import_memory,
+        'embeddings': import_embedding,
+        'vectorstores': import_vectorstore,
+        'documentloaders': import_documentloader,
+        'textsplitters': import_textsplitter,
+        'utilities': import_utility,
+        'output_parsers': import_output_parser,
+        'retrievers': import_retriever,
+        'autogenRoles': import_autogenRoles,
+        'inputOutput': import_inputoutput,
+    }
+    if _type == 'llms':
+        key = 'contribute' if name in chat_models.__all__ else 'chat' if 'chat' in name.lower(
+        ) else 'llm'
+        loaded_func = func_dict[_type][key]  # type: ignore
+    else:
+        loaded_func = func_dict[_type]
+
+    return loaded_func(name)
+
+
+def import_inputoutput(input_output: str) -> Any:
+    """Import output parser from output parser name"""
+    return import_module(f'from bisheng_langchain.input_output import {input_output}')
+
+
+def import_output_parser(output_parser: str) -> Any:
+    """Import output parser from output parser name"""
+    return import_module(f'from langchain.output_parsers import {output_parser}')
+
+
+def import_chat_llm(llm: str) -> BaseChatModel:
+    """Import chat llm from llm name"""
+    return import_class(f'langchain.chat_models.{llm}')
+
+
+def import_chain_contribute_llm(llm: str) -> BaseChatModel:
+    """Import chat llm from llm name"""
+    return import_class(f'bisheng_langchain.chat_models.{llm}')
+
+
+def import_retriever(retriever: str) -> Any:
+    """Import retriever from retriever name"""
+    return import_module(f'from langchain.retrievers import {retriever}')
+
+
+def import_autogenRoles(autogen: str) -> Any:
+    return import_module(f'from bisheng_langchain.autogen_role import {autogen}')
+
+
+def import_memory(memory: str) -> Any:
+    """Import memory from memory name"""
+    return import_module(f'from langchain.memory import {memory}')
+
+
+def import_prompt(prompt: str) -> Type[PromptTemplate]:
+    """Import prompt from prompt name"""
+    if prompt == 'ZeroShotPrompt':
+        return import_class('langchain.prompts.PromptTemplate')
+    return import_class(f'langchain.prompts.{prompt}')
+
+
+def import_toolkit(toolkit: str) -> Any:
+    """Import toolkit from toolkit name"""
+    return import_module(f'from langchain.agents.agent_toolkits import {toolkit}')
+
+
+def import_agent(agent: str) -> Agent:
+    """Import agent from agent name"""
+    # check for custom agent
+    from bisheng_langchain import agents
+    if agent in agents.__all__:
+        return import_class(f'bisheng_langchain.agents.{agent}')
+    return import_class(f'langchain.agents.{agent}')
+
+
+def import_llm(llm: str) -> BaseLanguageModel:
+    """Import llm from llm name"""
+    return import_class(f'langchain.llms.{llm}')
+
+
+def import_tool(tool: str) -> BaseTool:
+    """Import tool from tool name"""
+    return import_class(f'langchain.tools.{tool}')
+
+
+def import_chain(chain: str) -> Type[Chain]:
+    """Import chain from chain name"""
+    from bisheng_langchain import chains
+    if chain in chains.__all__:
+        return import_class(f'bisheng_langchain.chains.{chain}')
+    return import_class(f'langchain.chains.{chain}')
+
+
+def import_embedding(embedding: str) -> Any:
+    """Import embedding from embedding name"""
+    from bisheng_langchain import embeddings
+    if embedding in embeddings.__all__:
+        return import_class(f'bisheng_langchain.embeddings.{embedding}')
+    return import_class(f'langchain.embeddings.{embedding}')
+
+
+def import_vectorstore(vectorstore: str) -> Any:
+    """Import vectorstore from vectorstore name"""
+    from bisheng_langchain import vectorstores
+    if vectorstore in vectorstores.__all__:
+        return import_class(f'bisheng_langchain.vectorstores.{vectorstore}')
+    return import_class(f'langchain.vectorstores.{vectorstore}')
+
+
+def import_documentloader(documentloader: str) -> Any:
+    """Import documentloader from documentloader name"""
+    from bisheng_langchain import document_loaders
+    if documentloader in document_loaders.__all__:
+        return import_class(f'bisheng_langchain.document_loaders.{documentloader}')
+    return import_class(f'langchain.document_loaders.{documentloader}')
+
+
+def import_textsplitter(textsplitter: str) -> Any:
+    """Import textsplitter from textsplitter name"""
+    from bisheng_langchain import text_splitter
+    if textsplitter in dir(text_splitter):
+        return import_class(f'bisheng_langchain.text_splitter.{textsplitter}')
+    return import_class(f'langchain.text_splitter.{textsplitter}')
+
+
+def import_utility(utility: str) -> Any:
+    """Import utility from utility name"""
+    if utility == 'SQLDatabase':
+        return import_class(f'langchain.sql_database.{utility}')
+    return import_class(f'langchain.utilities.{utility}')

bisheng_langchain/retrievers/ensemble.py

@@ -67,6 +67,7 @@ class EnsembleRetriever(BaseRetriever):
         query: str,
         *,
         run_manager: AsyncCallbackManagerForRetrieverRun,
+        **kwagrs: Any,
     ) -> List[Document]:
         """
         Asynchronously get the relevant documents for a given query.
@@ -79,7 +80,7 @@ class EnsembleRetriever(BaseRetriever):
         """
 
         # Get fused result of the retrievers.
-        fused_documents = await self.arank_fusion(query, run_manager)
+        fused_documents = await self.arank_fusion(query, run_manager, **kwagrs)
 
         return fused_documents
 

bisheng_langchain/vectorstores/elastic_keywords_search.py

@@ -228,7 +228,7 @@ class ElasticKeywordsSearch(VectorStore, ABC):
         # llm or jiaba extract keywords
         if self.llm_chain:
             keywords_str = self.llm_chain.run(query)
-            print('keywords_str:', keywords_str)
+            print('llm search keywords:', keywords_str)
             try:
                 keywords = eval(keywords_str)
                 if not isinstance(keywords, list):
@@ -238,6 +238,7 @@ class ElasticKeywordsSearch(VectorStore, ABC):
                 keywords = jieba.analyse.extract_tags(query, topK=10, withWeight=False)
         else:
             keywords = jieba.analyse.extract_tags(query, topK=10, withWeight=False)
+            print('jieba search keywords:', keywords)
         match_query = {'bool': {must_or_should: []}}
         for key in keywords:
             match_query['bool'][must_or_should].append({query_strategy: {'text': key}})

{bisheng_langchain-0.3.0rc0.dist-info → bisheng_langchain-0.3.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: bisheng-langchain
-Version: 0.3.0rc0
+Version: 0.3.1
 Summary: bisheng langchain modules
 Home-page: https://github.com/dataelement/bisheng
 Author: DataElem

{bisheng_langchain-0.3.0rc0.dist-info → bisheng_langchain-0.3.1.dist-info}/RECORD

@@ -28,7 +28,7 @@ bisheng_langchain/chains/router/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm
 bisheng_langchain/chains/router/multi_rule.py,sha256=BiFryj3-7rOxfttD-MyOkKWLCSGB9LVYd2rjOsIfQC8,375
 bisheng_langchain/chains/router/rule_router.py,sha256=R2YRUnwn7s_7DbsSn27uPn4cIV0D-5iXEORXir0tNGM,1835
 bisheng_langchain/chat_models/__init__.py,sha256=4-HTLE_SXO4hmNJu6yQxiQKBt2IFca_ezllVBLmvbEE,635
-bisheng_langchain/chat_models/host_llm.py,sha256=sBu_Vg-r7z6IJUV8Etwll4JTG5OvET-IXH7PZw8Ijrc,23152
+bisheng_langchain/chat_models/host_llm.py,sha256=35_jTdUm85mk-t2MARZYGC8dIPVtf5XXlGfFE6hQ1Gc,23153
 bisheng_langchain/chat_models/minimax.py,sha256=JLs_f6vWD9beZYUtjD4FG28G8tZHrGUAWOwdLIuJomw,13901
 bisheng_langchain/chat_models/proxy_llm.py,sha256=wzVBZik9WC3-f7kyQ1eu3Ooibqpcocln08knf5lV1Nw,17082
 bisheng_langchain/chat_models/qwen.py,sha256=W73KxDRQBUZEzttEM4K7ZzPqbN-82O6YQmpX-HB_wZU,19971
@@ -49,7 +49,7 @@ bisheng_langchain/document_loaders/custom_kv.py,sha256=xWUPhcr1hjbdya4zgEHG4Fl0s
 bisheng_langchain/document_loaders/elem_html.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 bisheng_langchain/document_loaders/elem_image.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 bisheng_langchain/document_loaders/elem_pdf.py,sha256=K-TXILGNFLFjavhun_MFbUF4t2_WGA3Z-kbnr75lmW8,22243
-bisheng_langchain/document_loaders/elem_unstrcutured_loader.py,sha256=bJQObxHnk8FaF8RUBkqODzgeikrZ8wdl_TQPa2oEoQo,5169
+bisheng_langchain/document_loaders/elem_unstrcutured_loader.py,sha256=Dcu70Wz4vdUpq9y2vWPWroRdq6JrlOJNERoFTnZO0oU,5336
 bisheng_langchain/document_loaders/universal_kv.py,sha256=dJF_GQGKBMUjB_kX9CSp7xZRhXgwVuGPbMIzJwPh-C0,4063
 bisheng_langchain/document_loaders/parsers/__init__.py,sha256=OOM_FJkwaU-zNS58fASw0TH8FNT6VXKb0VrvisgdrII,171
 bisheng_langchain/document_loaders/parsers/ellm_client.py,sha256=B4Dea8xXXnGvB9j2OXv53HILNUmnWeNJz9ssNM-2fLM,1760
@@ -65,33 +65,35 @@ bisheng_langchain/embeddings/interface/__init__.py,sha256=GNY3tibpRxpAdAfSvQmXBK
 bisheng_langchain/embeddings/interface/types.py,sha256=VdurbtsnjCPdlOjPFcK2Mg6r9bJYYHb3tepvkk-y3nM,461
 bisheng_langchain/embeddings/interface/wenxin.py,sha256=5d9gI4enmfkD80s0FHKiDt33O0mwM8Xc5WTubnMUy8c,3104
 bisheng_langchain/gpts/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-bisheng_langchain/gpts/assistant.py,sha256=KCYPU1Bs4GtWcLk9Ya2NuQyXE0Twn7-92eSBTIzpq7I,5083
-bisheng_langchain/gpts/auto_optimization.py,sha256=Vf3zzYEpVf916dYt4RV9E1uw4vTXjE7ZXogUIdxjHYU,3786
-bisheng_langchain/gpts/auto_tool_selected.py,sha256=25lFLadqQ36t63EKMEF3zJOG_jkoRB9IfP5eRkY1JZo,1777
-bisheng_langchain/gpts/load_tools.py,sha256=C7tlRLy4wAArr9qtkRl9dW6QXdspLLbcv0UvulW9A9U,6345
+bisheng_langchain/gpts/assistant.py,sha256=OEB9HA-FpNKGOviVoafld4MHDIEQS5u9AU0QlMuJ-po,5344
+bisheng_langchain/gpts/auto_optimization.py,sha256=WNsC19rgvuDYQlSIaYThq5RqCbuobDbzCwAJW4Ksw0c,3626
+bisheng_langchain/gpts/auto_tool_selected.py,sha256=21WETf9o0YS-QEBwv3mmZRObKWszefQkXEqAA6KzoaM,1582
+bisheng_langchain/gpts/load_tools.py,sha256=LiiK1OqFu7Ki-F_Rhfi1rgp0wBQCSrTDdqsgwciTOIU,8099
 bisheng_langchain/gpts/message_types.py,sha256=7EJOx62j9E1U67jxWgxE_I7a8IjAvvKANknXkD2gFm0,213
 bisheng_langchain/gpts/utils.py,sha256=t3YDxaJ0OYd6EKsek7PJFRYnsezwzEFK5oVU-PRbu5g,6671
 bisheng_langchain/gpts/agent_types/__init__.py,sha256=bg0zlTYGfNXoSBqcICHlzNpVQbejMYeyji_dzvP5qQ0,261
-bisheng_langchain/gpts/agent_types/llm_functions_agent.py,sha256=ynFHXuaqABeiKvhcetOFQPyQMlNtEAdtTccoIwiJbGQ,8419
-bisheng_langchain/gpts/prompts/__init__.py,sha256=IfuoxVpsSLKJtDx0aJbRgnSZYZr_kDPsL92CvefzF-s,568
+bisheng_langchain/gpts/agent_types/llm_functions_agent.py,sha256=IXg5u8dSk-FcLvjrvvLcN5revGccXylXkD73ZWhaDWs,8715
+bisheng_langchain/gpts/prompts/__init__.py,sha256=pOnXvk6_PjqAoLrh68sI9o3o6znKGxoLMVFP-0XTCJo,704
+bisheng_langchain/gpts/prompts/assistant_prompt_base.py,sha256=Q69qWxtwaeGPPwMWXQ44OsQPg7wHB7nYT17wbqQe3kM,57
+bisheng_langchain/gpts/prompts/assistant_prompt_cohere.py,sha256=GLQ77oXqSlE7Xes2ObsFsNon5nOJOCRhQOKE5bUpgaI,2421
 bisheng_langchain/gpts/prompts/assistant_prompt_opt.py,sha256=TZsRK4XPMrUhGg0PoMyiE3wE-aG34UmlVflkCl_c0QI,4151
-bisheng_langchain/gpts/prompts/base_prompt.py,sha256=v2eO0c6RF8e6MtGdleHs5B4YTkikg6IZUuBvL2zvyOI,55
 bisheng_langchain/gpts/prompts/breif_description_prompt.py,sha256=w4A5et0jB-GkxEMQBp4i6GKX3RkVeu7NzWEjOZZAicM,5336
-bisheng_langchain/gpts/prompts/opening_dialog_prompt.py,sha256=U6SDslWuXAB1ZamLZVujpEjAY8L244IZfD2qFVRTzPM,5962
+bisheng_langchain/gpts/prompts/opening_dialog_prompt.py,sha256=VVF0JLHtetupVB0kabiFHWDHlQaa4nFLcbYXgIBA3nw,5965
 bisheng_langchain/gpts/prompts/select_tools_prompt.py,sha256=AyvVnrLEsQy7RHuGTPkcrMUxgA98Q0TzF-xweoc7GyY,1400
 bisheng_langchain/gpts/tools/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-bisheng_langchain/gpts/tools/api_tools/__init__.py,sha256=mrmTV5bT5R1mEx9hbMAWKzNAAC4EL6biNn53dx5lYsc,1593
-bisheng_langchain/gpts/tools/api_tools/base.py,sha256=t0gFRiXD-2InCHNDU_T1H0eSMXsWP_2sHBeLJ0bZOkc,3594
-bisheng_langchain/gpts/tools/api_tools/flow.py,sha256=u1_ASWlCcZarKR-293kACB_qQ1RzJuzPC3YZSl2JR-E,1814
-bisheng_langchain/gpts/tools/api_tools/macro_data.py,sha256=rlFNhjJ3HEHfWeW9Wqb27eeF1Q1Qmd2SA8VfgUK4ACs,19270
-bisheng_langchain/gpts/tools/api_tools/sina.py,sha256=tY4MXHNBEUiBN1wld2un_w7nHW_njXV7haLo3sgpDf0,9502
+bisheng_langchain/gpts/tools/api_tools/__init__.py,sha256=CkEjgIFM4GIv86V1B7SsFLaB6M86c54QuO8wIRizUZ8,1608
+bisheng_langchain/gpts/tools/api_tools/base.py,sha256=fWQSDIOVb4JZrtJ9ML9q2ycsAa-_61gXTD0MT19J1LM,3618
+bisheng_langchain/gpts/tools/api_tools/flow.py,sha256=ot2YAYgQGWgUpb2nCECAmpqHY6m0SgzwkupF9kDT3lU,2461
+bisheng_langchain/gpts/tools/api_tools/macro_data.py,sha256=FyG-qtl2ECS1CDKt6olN0eDTDM91d-UvDkMDBiVLgYQ,27429
+bisheng_langchain/gpts/tools/api_tools/openapi.py,sha256=3MlIpzR_NhZogaA-zbH6fnT_KUMm10NnbtDYcxKedS8,3907
+bisheng_langchain/gpts/tools/api_tools/sina.py,sha256=GGA4ZYvNEpqBZ_l8MUYqgkI8xZe9XcGa9-KlHZVqr6I,9542
 bisheng_langchain/gpts/tools/api_tools/tianyancha.py,sha256=abDAz-yAH1-2rKiSmZ6TgnrNUnpgAZpDY8oDiWfWapc,6684
 bisheng_langchain/gpts/tools/bing_search/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 bisheng_langchain/gpts/tools/bing_search/tool.py,sha256=v_VlqcMplITA5go5qWA4qZ5p43E1-1s0bzmyY7H0hqY,1710
 bisheng_langchain/gpts/tools/calculator/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 bisheng_langchain/gpts/tools/calculator/tool.py,sha256=iwGPE7jvxZg_jUL2Aq9HHwnRJrF9-ongwrsBX6uk1U0,705
 bisheng_langchain/gpts/tools/code_interpreter/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-bisheng_langchain/gpts/tools/code_interpreter/tool.py,sha256=PGipxd-qtW31GonRGfGow7nylI-osSnmBsvEJDlMUCE,8717
+bisheng_langchain/gpts/tools/code_interpreter/tool.py,sha256=1VLkgngRR0k8YjA4eYkfPd1E7fD29tMKpqtCtn7WwYE,11443
 bisheng_langchain/gpts/tools/dalle_image_generator/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 bisheng_langchain/gpts/tools/dalle_image_generator/tool.py,sha256=mhxdNNhBESjbOy30Rnp6hQhnrV4evQpv-B1fFXcU-68,7528
 bisheng_langchain/gpts/tools/get_current_time/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -99,16 +101,49 @@ bisheng_langchain/gpts/tools/get_current_time/tool.py,sha256=3uvk7Yu07qhZy1sBrFM
 bisheng_langchain/input_output/__init__.py,sha256=sW_GB7MlrHYsqY1Meb_LeimQqNsMz1gH-00Tqb2BUyM,153
 bisheng_langchain/input_output/input.py,sha256=I5YDmgbvvj1o2lO9wi8LE37wM0wP5jkhUREU32YrZMQ,1094
 bisheng_langchain/input_output/output.py,sha256=6U-az6-Cwz665C2YmcH3SYctWVjPFjmW8s70CA_qphk,11585
+bisheng_langchain/rag/__init__.py,sha256=gGa3hx0HjA_11FxbnR3P0C0y_OwcZM-sYTHdAajmylk,102
+bisheng_langchain/rag/bisheng_rag_pipeline.py,sha256=neoBK3TtuQ07_WeuJCzYlvtsDQNepUa_68NT8VCgytw,13749
+bisheng_langchain/rag/bisheng_rag_pipeline_v2.py,sha256=FWmhOSLAKWqqhdj55Y6WoFJ5GYwK8fJdN3o-mqPmabI,15898
+bisheng_langchain/rag/bisheng_rag_pipeline_v2_cohere_raw_prompting.py,sha256=Gql3IXaSTtW8WaC7wLLid0TURXoF0hRqAgl3GAYBBCo,16816
+bisheng_langchain/rag/bisheng_rag_tool.py,sha256=X7mhqYHapLlqfpu90BaW-C3aDeW_PENIci9lNyQv-tI,11971
+bisheng_langchain/rag/extract_info.py,sha256=emrdvzU9EfdTQU_IBF4RkwaTXmYLk1MOJgdfYhw6VtA,1753
+bisheng_langchain/rag/run_qa_gen_web.py,sha256=-fIvHNnD3lD0iNU5m0Me1GDwRjlcsB8tE5RnPtFRG2s,1840
+bisheng_langchain/rag/run_rag_evaluate_web.py,sha256=a9vMhq-ZhEiHHr43uKUzKtjdk280uAP_UHQW_eOaQMw,2224
+bisheng_langchain/rag/utils.py,sha256=ZR3BwjjLX4XT49aRDv-pf8Z4Mqcbmt2_qbDSwXfDfso,6361
+bisheng_langchain/rag/config/baseline.yaml,sha256=cFkfVpuNtmexYkQxXbqbcqQ4eDhx81kmIXf8T_rc_Oo,2266
+bisheng_langchain/rag/config/baseline_caibao.yaml,sha256=uU2HSlPXB8uLq9Fh03wENs06STdMA9CS6ry3X-Wkn-E,2174
+bisheng_langchain/rag/config/baseline_caibao_knowledge_v2.yaml,sha256=KrkMmHu-myQ8u4FlydXLwsppD4ucQGAYYtGGA0HCJzE,2812
+bisheng_langchain/rag/config/baseline_caibao_v2.yaml,sha256=WY9wek70eeBhhyiNmkfyrffxhTfBntqPyLMRmoqM_vE,2846
+bisheng_langchain/rag/config/baseline_demo_v2.yaml,sha256=hCa7bZMCcOWpu6bsAFX4DU1b7MV4eN2phyzvZzFZKkU,2382
+bisheng_langchain/rag/config/baseline_s2b_mix.yaml,sha256=rkPfzU2-mvjRrZ0zMHaQsncPhq8DrdvVFsw4Sg_jeKc,2398
+bisheng_langchain/rag/config/baseline_v2.yaml,sha256=yUMlK9CbeIqIrb4iJj14BaqH7JJ15QdUIf1EHd7T_R8,2350
+bisheng_langchain/rag/init_retrievers/__init__.py,sha256=qpLLAuqZPtumTlJj17Ie5AbDDmiUiDxYefg_pumqu-c,218
+bisheng_langchain/rag/init_retrievers/baseline_vector_retriever.py,sha256=oRKZZpxlLQAtsubIcAXeXpf1a9h6Pt6uOtNTLeD2jps,2362
+bisheng_langchain/rag/init_retrievers/keyword_retriever.py,sha256=NRT0fBx6HFR7j9IbRl_NBuqF7hnL-9v5GCqHpgnrfPQ,2523
+bisheng_langchain/rag/init_retrievers/mix_retriever.py,sha256=Whxq4kjNPLsxnHcVo60usdFFwLTCD-1jO38q08LXkVQ,4653
+bisheng_langchain/rag/init_retrievers/smaller_chunks_retriever.py,sha256=RQ7QLEOOhBrkw-EimXVJqIGa96D-KkNDik2h9hzg9fU,3805
+bisheng_langchain/rag/prompts/__init__.py,sha256=IUCq9gzqGQN_6IDk0D_F5t3mOUI_KbmSzYnnXoX4VKE,223
+bisheng_langchain/rag/prompts/extract_key_prompt.py,sha256=THdcwolRzaKkddgEhabUDytn-caHs9UwECPkPxijWAs,1456
+bisheng_langchain/rag/prompts/prompt.py,sha256=HPQ4mnMXDL9YsgBqK4S1M2sE6icq2uDL17KKgvEkovI,1925
+bisheng_langchain/rag/prompts/prompt_cohere.py,sha256=FhW-RCIVnIfKtEqV_-7aQav1riIJmA0jV2hGxCSoysk,8151
+bisheng_langchain/rag/qa_corpus/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+bisheng_langchain/rag/qa_corpus/qa_generator.py,sha256=ZdipOLBd16XfKXygq1JeUx4iv3tfUsprk1HNuC5hcIk,6092
+bisheng_langchain/rag/rerank/__init__.py,sha256=lcMRa_o5HGNnN9Hn6zHCimYq02OuD9pRAg6eFPZNgR8,72
+bisheng_langchain/rag/rerank/rerank.py,sha256=RoEwFFb4t4l0aZQmE2-HHHfWIzNLKe9NuFDIoYoWz5g,1836
+bisheng_langchain/rag/rerank/rerank_benchmark.py,sha256=OU9bh7dQ0-faw4hpz6BPEtmbQEhISTekvRiWpksOfCc,4490
+bisheng_langchain/rag/scoring/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+bisheng_langchain/rag/scoring/llama_index_score.py,sha256=Xh8YFT14JGxE3OL2x4JSRa8nGlOK5DlWlsrzbmA2gu8,3001
+bisheng_langchain/rag/scoring/ragas_score.py,sha256=ClVoRg1O7s5OMcsB2E5UbVHj_ErQ3WKr9hao8BcK6NA,7110
 bisheng_langchain/retrievers/__init__.py,sha256=XqBeNyPyNCJf-SzNBiFlkxtjrtHUFTTi5pe2yPyOKrA,210
-bisheng_langchain/retrievers/ensemble.py,sha256=nLsTKpJmaigrECCWzrvDUwhE-qs9Mg7gPRXfPo5qFMI,5942
+bisheng_langchain/retrievers/ensemble.py,sha256=umjBaZYBEdhJ2F7GlzQgXVLYjSfpybTptiJJbUgryZE,5975
 bisheng_langchain/retrievers/mix_es_vector.py,sha256=dSrrsuMPSgGiu181EOzACyIKiDXR0qNBQz_914USD3E,4465
 bisheng_langchain/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 bisheng_langchain/utils/requests.py,sha256=vWGKyNTxApVeaVdKxqACfIT1Q8wMy-jC3kUv2Ce9Mzc,8688
 bisheng_langchain/vectorstores/__init__.py,sha256=zCZgDe7LyQ0iDkfcm5UJ5NxwKQSRHnqrsjx700Fy11M,213
-bisheng_langchain/vectorstores/elastic_keywords_search.py,sha256=JV_GM40cYx0PtPPvH2JYxtsMV0psSW2CDKagpR4M_0o,13286
+bisheng_langchain/vectorstores/elastic_keywords_search.py,sha256=Pm1rS50GJ0HWbjBsFDgs28SVuVbjGSRPOor6yJlnE7w,13347
 bisheng_langchain/vectorstores/milvus.py,sha256=lrnezKnYXhyH5M1g3a-Mcwpj9mwzAj44TKmzyUXlQYY,36297
 bisheng_langchain/vectorstores/retriever.py,sha256=hj4nAAl352EV_ANnU2OHJn7omCH3nBK82ydo14KqMH4,4353
-bisheng_langchain-0.3.0rc0.dist-info/METADATA,sha256=-wUMNJfiSIK6i3D9pP74PdmDKuLtDcrW3BcNpNzqv1A,2414
-bisheng_langchain-0.3.0rc0.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-bisheng_langchain-0.3.0rc0.dist-info/top_level.txt,sha256=Z6pPNyCo4ihyr9iqGQbH8sJiC4dAUwA_mAyGRQB5_Fs,18
-bisheng_langchain-0.3.0rc0.dist-info/RECORD,,
+bisheng_langchain-0.3.1.dist-info/METADATA,sha256=uorWxDbAnQkGAmbh0c-ZmAnTf5MgzyhcY8hLKNDIp-Q,2411
+bisheng_langchain-0.3.1.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
+bisheng_langchain-0.3.1.dist-info/top_level.txt,sha256=Z6pPNyCo4ihyr9iqGQbH8sJiC4dAUwA_mAyGRQB5_Fs,18
+bisheng_langchain-0.3.1.dist-info/RECORD,,

bisheng_langchain/gpts/prompts/base_prompt.py

@@ -1 +0,0 @@
-DEFAULT_SYSTEM_MESSAGE = "You are a helpful assistant."