Datasets:

DanCip

/

github-issues-negatives

like 0

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python final_chain = ConversationalRetrievalChain.from_llm( llm=llm, retriever=retriever, memory=memory, verbose=True ) ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description I am using MongoDBChatMessageHistory and it is working as expected in terms of functionality but I would expect the history record to be saved in an object in MongoDB rather than a string to aid with readability in the database. (as an aside I am not sure why the fields are not snake or camelcase...)  ### System Info ``` langchain==0.1.13 langchain-community==0.0.29 langchain-core==0.1.36 langchain-mongodb==0.1.3 langchain-openai==0.1.1 langchain-text-splitters==0.0.1 langchainhub==0.1.15 ```

MongoDBChatMessageHistory saves History as a string rather than an object

https://api.github.com/repos/langchain-ai/langchain/issues/20486/comments

2024-04-15T18:56:38Z

2024-07-23T16:12:02Z

https://github.com/langchain-ai/langchain/issues/20486

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code The following code: ``` from langchain_community.chat_models import ChatCohere from dotenv import load_dotenv import os load_dotenv() COHERE_API_KEY= os.getenv("COHERE_API_KEY") chat_model = ChatCohere(cohere_api_key=COHERE_API_KEY, model="command-r") test = await chat_model.ainvoke("test") print(test) ``` ### Error Message and Stack Trace (if applicable) --------------------------------------------------------------------------- AttributeError Traceback (most recent call last) Cell In[4], [line 8](vscode-notebook-cell:?execution_count=4&line=8) [6](vscode-notebook-cell:?execution_count=4&line=6) COHERE_API_KEY= os.getenv("COHERE_API_KEY") [7](vscode-notebook-cell:?execution_count=4&line=7) chat_model = ChatCohere(cohere_api_key=COHERE_API_KEY, model="command-r",offline=False) ----> [8](vscode-notebook-cell:?execution_count=4&line=8) test = await chat_model.ainvoke("test") [9](vscode-notebook-cell:?execution_count=4&line=9) print(test) File [~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:175](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:175), in BaseChatModel.ainvoke(self, input, config, stop, **kwargs) [166](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:166) async def ainvoke( [167](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:167) self, [168](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:168) input: LanguageModelInput, (...) [172](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:172) **kwargs: Any, [173](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:173) ) -> BaseMessage: [174](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:174) config = ensure_config(config) --> [175](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:175) llm_result = await self.agenerate_prompt( [176](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:176) [self._convert_input(input)], [177](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:177) stop=stop, [178](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:178) callbacks=config.get("callbacks"), [179](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:179) tags=config.get("tags"), [180](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:180) metadata=config.get("metadata"), [181](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:181) run_name=config.get("run_name"), [182](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:182) run_id=config.pop("run_id", None), [183](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:183) **kwargs, [184](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:184) ) [185](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:185) return cast(ChatGeneration, llm_result.generations[0][0]).message File [~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:566](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:566), in BaseChatModel.agenerate_prompt(self, prompts, stop, callbacks, **kwargs) [558](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:558) async def agenerate_prompt( [559](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:559) self, [560](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:560) prompts: List[PromptValue], (...) [563](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:563) **kwargs: Any, [564](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:564) ) -> LLMResult: [565](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:565) prompt_messages = [p.to_messages() for p in prompts] --> [566](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:566) return await self.agenerate( [567](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:567) prompt_messages, stop=stop, callbacks=callbacks, **kwargs [568](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:568) ) File [~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:526](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:526), in BaseChatModel.agenerate(self, messages, stop, callbacks, tags, metadata, run_name, run_id, **kwargs) [513](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:513) if run_managers: [514](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:514) await asyncio.gather( [515](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:515) *[ [516](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:516) run_manager.on_llm_end( (...) [524](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:524) ] [525](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:525) ) --> [526](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:526) raise exceptions[0] [527](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:527) flattened_outputs = [ [528](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:528) LLMResult(generations=[res.generations], llm_output=res.llm_output) # type: ignore[list-item, union-attr] [529](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:529) for res in results [530](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:530) ] [531](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:531) llm_output = self._combine_llm_outputs([res.llm_output for res in results]) # type: ignore[union-attr] File [~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:707](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:707), in BaseChatModel._agenerate_with_cache(self, messages, stop, run_manager, **kwargs) [705](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:705) else: [706](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:706) if inspect.signature(self._agenerate).parameters.get("run_manager"): --> [707](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:707) result = await self._agenerate( [708](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:708) messages, stop=stop, run_manager=run_manager, **kwargs [709](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:709) ) [710](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:710) else: [711](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py:711) result = await self._agenerate(messages, stop=stop, **kwargs) File [~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:242](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:242), in ChatCohere._agenerate(self, messages, stop, run_manager, **kwargs) [240](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:240) generation_info = None [241](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:241) if hasattr(response, "documents"): --> [242](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:242) generation_info = self._get_generation_info(response) [243](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:243) return ChatResult( [244](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:244) generations=[ [245](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:245) ChatGeneration(message=message, generation_info=generation_info) [246](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:246) ] [247](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:247) ) File [~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:194](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:194), in ChatCohere._get_generation_info(self, response) [187](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:187) def _get_generation_info(self, response: Any) -> Dict[str, Any]: [188](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:188) """Get the generation info from cohere API response.""" [189](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:189) return { [190](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:190) "documents": response.documents, [191](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:191) "citations": response.citations, [192](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:192) "search_results": response.search_results, [193](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:193) "search_queries": response.search_queries, --> [194](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:194) "token_count": response.token_count, [195](https://file+.vscode-resource.vscode-cdn.net/home/jfuehne/Desktop/AI/Code/chatbotlocal/~/Desktop/AI/.venv/lib/python3.11/site-packages/langchain_community/chat_models/cohere.py:195) } AttributeError: 'NonStreamedChatResponse' object has no attribute 'token_count' ### Description I realized that recently, all of my code for Cohere through langchain has broken due to API errors. Is this possibly due to this change? https://docs.cohere.com/changelog/python-sdk-v520-release currently, it fails on even the most basic test of just using invoke on the chat model ### System Info langchain==0.1.16 langchain-anthropic==0.1.4 langchain-community==0.0.32 langchain-core==0.1.42 langchain-text-splitters==0.0.1 langchainhub==0.1.15 cohere==5.2.5 Platform: Linux python version: 3.11.8

AttributeError: 'NonStreamedChatResponse' object has no attribute 'token_count'

https://api.github.com/repos/langchain-ai/langchain/issues/20484/comments

2024-04-15T18:20:49Z

2024-04-16T16:51:51Z

https://github.com/langchain-ai/langchain/issues/20484

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code Failed to batch ingest runs: LangSmithError('Failed to POST https://api.smith.langchain.com/runs/batch in LangSmith API. HTTPError(\'403 Client Error: Forbidden for url: https://api.smith.langchain.com/runs/batch\', \'{"detail":"Forbidden"}\')') ### Description I am using crewai along with langsmith and the error looks like a server side error. ### System Info platform windows python 3.11.4

Failed to batch ingest runs: LangSmithError('Failed to POST https://api.smith.langchain.com/runs/batch in LangSmith API. HTTPError(\'403 Client Error: Forbidden for url: https://api.smith.langchain.com/runs/batch\', \'{"detail":"Forbidden"}\')')

https://api.github.com/repos/langchain-ai/langchain/issues/20479/comments

2024-04-15T16:05:48Z

2024-08-02T15:00:06Z

https://github.com/langchain-ai/langchain/issues/20479

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```def load_documents(): texts = [] all_items = os.listdir("files/") for _, item in enumerate(all_items): print("Processing file: {}".format(item)) # Generate a unique id for each document unique_id = str(uuid.uuid4()) text_splitter = RecursiveCharacterTextSplitter(chunk_size=200, chunk_overlap=0, add_start_index=True) loader = PDFMinerLoader("files/{}".format(item), extract_images=True) docs = loader.load_and_split(text_splitter) # Add unique id and line range to each document for doc in docs: doc.metadata["unique_id"] = unique_id texts.extend(docs) return texts ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description Langchain text splitters needs a functionality to store chunk metadata. Either a start and end line number in pdf or coordinates in pdf. Currently its not possible in the text splitters. There are 2 options for this to store the chunk information. 1. Store from and to line. { "chunk": "reversible llowing ......", "pageNumber": 1, "fromLine": 27, "toLine": 40, "documentId": "8a385d38-63ed-4821-a5c2-2e7309d5a256" }, 2. storing the coordinates of a chunk { "text": "\n Coach : the ai chatbot - We will have several coaches with different names, each with a unique", "coordinates": [ { "x": 54, "y": 130.7924798925 }, { "x": 512.654318754038, "y": 141.7924801425 } ], "pageSize": { "width": 612, "height": 792 }, "pageNumber": 1 }, Looking for some help here. Thanks! ### System Info $ pip freeze | grep langchain langchain==0.1.9 langchain-community==0.0.24 langchain-core==0.1.40 langchain-openai==0.0.8 windows python 3.11

Langchain text splitters need a functionality to store chunk metadata

https://api.github.com/repos/langchain-ai/langchain/issues/20470/comments

2024-04-15T13:15:14Z

2024-07-24T16:08:31Z

https://github.com/langchain-ai/langchain/issues/20470

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code agent = planner.create_openapi_agent( api_spec=api_spec, requests_wrapper=requests_wrapper, llm=llm, allow_dangerous_requests=True, ) ### Error Message and Stack Trace (if applicable) _No response_ ### Description i want the tool to have full HTTP capabilities and not only GET and POST ### System Info latest langchain python installation which is 0.1.16

OpenAPI agent does not support PATCH and PUT and DELETE, only GET and POST

https://api.github.com/repos/langchain-ai/langchain/issues/20469/comments

2024-04-15T12:40:10Z

2024-06-22T00:13:06Z

https://github.com/langchain-ai/langchain/issues/20469

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code doc_list_1 = [ "I like apples", "I like oranges", "Apples and oranges are fruits", ] ret2 = BM25Retriever.from_texts( doc_list_1, metadatas=[{"source": 1}] * len(doc_list_1) ) retriever = EnsembleRetriever( retrievers=[ret2, retriever], weights=[0.5, 0.5] ) ### Error Message and Stack Trace (if applicable) _No response_ ### Description i am trying to use ensemble retriever containing bm25 and opensearch i am getting the below error :- "1 validation error for Document\npage_content\n str type expected (type=type_error.str)" ### System Info langchain==0.1.0 langchain-community==0.0.9 langchain-core==0.1.7 langchainhub==0.1.14

not able to run BM25 retriever

https://api.github.com/repos/langchain-ai/langchain/issues/20466/comments

2024-04-15T10:47:21Z

2024-07-22T16:08:56Z

https://github.com/langchain-ai/langchain/issues/20466

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code from langchain.sql_database import SQLDatabase from langchain.prompts import ChatPromptTemplate, MessagesPlaceholder,PromptTemplate from langchain.tools import BaseTool from langchain.tools.render import format_tool_to_openai_function from langchain.schema.runnable import Runnable,RunnableLambda,RunnableParallel from langchain.chat_models import ChatOpenAI from langchain.agents.format_scratchpad import format_to_openai_function_messages from langchain.agents.output_parsers import OpenAIFunctionsAgentOutputParser from langchain.agents import AgentExecutor from pydantic import BaseModel, Field import os from secret_key import openapi_key from sqlalchemy import create_engine import constants from datetime import datetime os.environ['OPENAI_API_KEY'] = openapi_key SQL_DML_CHAIN_PROMPT = """You are expert in SQLITE. Your main objective is to construct Data manipulation SQLITE query give the user question: {user_question}. You need to construct the Data manipulation SQLITE query for the following Database Schema: {table_info} Only Output the final SQL-Query and nothing else. SQL-Query:""" prompt = PromptTemplate(template = SQL_DML_CHAIN_PROMPT,input_variables = ['user_question','table_info']) from urllib.parse import quote_plus server_name = constants.server_name database_name = constants.database_name username = constants.username password = constants.password encoded_password = quote_plus(password) connection_uri = f"mssql+pyodbc://{username}:{encoded_password}@{server_name}/{database_name}?driver=ODBC+Driver+17+for+SQL+Server" engine = create_engine(connection_uri) model_name = "get-3.5-turbo-16k" db = SQLDatabase(engine, view_support=True, include_tables=['PAY_transaction_settingallow', 'PAY_mst_allowance','employee_details'],sample_rows_in_table_info=5) sql_dml_chain = RunnableParallel ({"user_question":lambda x:x["user_question"], "table_info":lambda _: db.get_table_info()}) |\ prompt |\ ChatOpenAI().bind(stop='SQL-Query:') |\ RunnableLambda(lambda x:x.content.strip().replace('```sql','')) agent_prompt = ChatPromptTemplate.from_messages( [ ("system", """ You are expert in SQL whose main objective is to mainpulate the Database for which you have been given access. You can use the tool `sql_db_manipulation` to interact with Database and mainpulate the database as per the user requirement. Wrapped column names: All column names should be wrapped in square brackets [] as delimiters. Use GETDATE() to get the current date and time instead of DATETIME('now'). For each record processed, retrieve the (employeeEuid) as 'euid' from the employee_details table where the employeeName matches, and retrieve the allowance ID (alw_id) from the pay_mst_allowance table where the alw_desc matches.no changes has to be made in "pay_mst_allowance" table and "employee_details" table. perform JOIN operation to fetch euid and alw_id from respective tables. Selected table: Specify PAY_transaction_settingallow as the table to update. Employee and allowance selection: Use the WHERE clause to filter employees based on employeeName and allowances based on alw_desc. Date handling: Maintain the logic for createDate, effect_date, and to_date using SQL Server functions. Currency: Assume the amount to be in rupees. Removed newlines: Write the query as a single string without newlines (\n). Ensure the query executes efficiently and without errors. """), ("user", "{input}"), MessagesPlaceholder(variable_name="agent_scratchpad"), ] ) class SQLDBMANIPULATION(BaseModel): user_query: str = Field(description='User question which will be translated to a Data Manipulation SQL Query and will be executed on the underlying database') class SQLDBMANIPULATIONTool(BaseTool): name = "sql_db_manipulation" description = "Use this tool to convert and execute DML queries given the user question" args_schema: type[SQLDBMANIPULATION] = SQLDBMANIPULATION sql_dml_chain: Runnable def _run( self, user_query: str ) -> str: """Use the tool.""" query = sql_dml_chain.invoke({"user_question":user_query}) query = query.replace("DATETIME('now')", "GETDATE()") db._execute(query) tools = [SQLDBMANIPULATIONTool(sql_dml_chain = sql_dml_chain)] llm_with_tools = ChatOpenAI().bind(functions=[format_tool_to_openai_function(t) for t in tools]) agent = ( { "input": lambda x: x["input"], "agent_scratchpad": lambda x: format_to_openai_function_messages( x["intermediate_steps"] ), } | agent_prompt | llm_with_tools | OpenAIFunctionsAgentOutputParser() ) agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True) ### Error Message and Stack Trace (if applicable) > Entering new AgentExecutor chain... Invoking: `sql_db_manipulation` with `{'user_query': "INSERT INTO PAY_transaction_settingallow ([euid], [alw_id], [createDate], [effect_date], [to_date], [amount], [currency]) SELECT ed.employeeEuid AS euid, pma.alw_id, GETDATE() AS createDate, '2024-03-01', '2024-03-31', 500, 'INR' FROM employee_details ed JOIN pay_mst_allowance pma ON ed.employeeName = 'jyothi' AND pma.alw_desc = 'Production Incentive'", 'user_question': 'insert Production incentive of 500 For jyothi for march 2024'}` Traceback (most recent call last): File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\sqlalchemy\engine\base.py", line 1969, in _exec_single_context self.dialect.do_execute( File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\sqlalchemy\engine\default.py", line 922, in do_execute cursor.execute(statement, parameters) pyodbc.ProgrammingError: ('42000', "[42000] [Microsoft][ODBC Driver 17 for SQL Server][SQL Server]'datetime' is not a recognized built-in function name. (195) (SQLExecDirectW)") The above exception was the direct cause of the following exception: Traceback (most recent call last): File "c:\Users\jyothi\Desktop\test\modifying.py", line 130, in <module> agent_executor.invoke({"input": " insert Production incentive of 500 For jyothi for march 2024 "}) File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\langchain\chains\base.py", line 87, in invoke return self( ^^^^^ File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\langchain\chains\base.py", line 310, in __call__ raise e File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\langchain\chains\base.py", line 304, in __call__ self._call(inputs, run_manager=run_manager) File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\langchain\agents\agent.py", line 1245, in _call next_step_output = self._take_next_step( ^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\langchain\agents\agent.py", line 1095, in _take_next_step observation = tool.run( ^^^^^^^^^ File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\langchain\tools\base.py", line 365, in run raise e File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\langchain\tools\base.py", line 339, in run else self._run(*tool_args, **tool_kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "c:\Users\jyothi\Desktop\test\modifying.py", line 98, in _run db._execute(query) File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\langchain\utilities\sql_database.py", line 411, in _execute cursor = connection.execute(text(command)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\sqlalchemy\engine\base.py", line 1416, in execute return meth( ^^^^^ File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\sqlalchemy\sql\elements.py", line 516, in _execute_on_connection return connection._execute_clauseelement( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\sqlalchemy\engine\base.py", line 1639, in _execute_clauseelement ret = self._execute_context( ^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\sqlalchemy\engine\base.py", line 1848, in _execute_context return self._exec_single_context( ^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\sqlalchemy\engine\base.py", line 1988, in _exec_single_context self._handle_dbapi_exception( File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\sqlalchemy\engine\base.py", line 2343, in _handle_dbapi_exception raise sqlalchemy_exception.with_traceback(exc_info[2]) from e File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\sqlalchemy\engine\base.py", line 1969, in _exec_single_context self.dialect.do_execute( File "C:\Users\jyothi\Desktop\test\gpttest\Lib\site-packages\sqlalchemy\engine\default.py", line 922, in do_execute cursor.execute(statement, parameters) sqlalchemy.exc.ProgrammingError: (pyodbc.ProgrammingError) ('42000', "[42000] [Microsoft][ODBC Driver 17 for SQL Server][SQL Server]'datetime' is not a recognized built-in function name. (195) (SQLExecDirectW)") [SQL: INSERT INTO PAY_transaction_settingallow ([euid], [alw_id], [createDate], [effect_date], [to_date], [amount], [currency]) SELECT ed.employeeEuid AS euid, pma.alw_id, datetime('now') AS createDate, '2024-03-01', '2024-03-31', 500, 'INR' FROM employee_details ed JOIN PAY_mst_allowance pma ON ed.employeeName = 'jyothi' AND pma.alw_desc = 'Production Incentive';] ### Description I'm using the agentExecutor from langhain to database manipulation to perform curd operation on the table, even after specifically mentioning to use GETDATE() it keep taking datetime('now') while executing the query i have shared the complete error while Invoking it write a correct query Invoking: `sql_db_manipulation` with `{'user_query': "INSERT INTO PAY_transaction_settingallow ([euid], [alw_id], [createDate], [effect_date], [to_date], [amount], [currency]) SELECT ed.employeeEuid AS euid, pma.alw_id, GETDATE() AS createDate, '2024-03-01', '2024-03-31', 500, 'INR' FROM employee_details ed JOIN pay_mst_allowance pma ON ed.employeeName = 'jyothi' AND pma.alw_desc = 'Production Incentive'", 'user_question': 'insert Production incentive of 500 For jyothi for march 2024'}` but while executing its taking [SQL: INSERT INTO PAY_transaction_settingallow ([euid], [alw_id], [createDate], [effect_date], [to_date], [amount], [currency]) SELECT ed.employeeEuid AS euid, pma.alw_id, datetime('now') AS createDate, '2024-03-01', '2024-03-31', 500, 'INR' FROM employee_details ed JOIN PAY_mst_allowance pma ON ed.employeeName = 'jyothi' AND pma.alw_desc = 'Production Incentive';] ### System Info os: windows python: 3.11 langchain: latest

'datetime' is not a recognized built-in function name. with angen_executor

https://api.github.com/repos/langchain-ai/langchain/issues/20465/comments

2024-04-15T10:44:40Z

2024-08-09T08:37:35Z

https://github.com/langchain-ai/langchain/issues/20465

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code Not applicable ### Error Message and Stack Trace (if applicable) _No response_ ### Description The following files trigger antivirus scans which report the presence of the Emf.Exploit.CVE_2017-3122-6335825-0 CVE: * docs/docs/integrations/document_loaders/example_data/fake.vsdx * libs/community/tests/examples/fake.vsdx These files were added in the https://github.com/langchain-ai/langchain/pull/16171 PR. Details on the scan results: https://www.virustotal.com/gui/file/3b02db67f312bfb1a0ac430673c372ec92eabfaf2888030161d7841ae2120f5f/detection Recommendation: remove the `visio/media/image2.emf` entry from the `fake.vsdx` archive. This is the file which triggers the CVE and it is not required for tests which use the archive. ### System Info System Information ------------------ > OS: Linux > OS Version: #26~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Mar 12 10:22:43 UTC 2 > Python Version: 3.11.4 (main, Jul 10 2023, 09:48:51) [GCC 11.3.0] Package Information ------------------- > langchain_core: 0.1.26 > langchain: 0.1.9 > langchain_community: 0.0.22 > langsmith: 0.1.5 > langchain_experimental: 0.0.52 > langchain_openai: 0.0.7 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve

community: test file triggers antivirus scan

https://api.github.com/repos/langchain-ai/langchain/issues/20456/comments

2024-04-15T08:50:57Z

2024-07-29T16:07:22Z

https://github.com/langchain-ai/langchain/issues/20456

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python from langchain_community.chat_models import PaiEasChatEndpoint from langchain_core.prompts import ChatPromptTemplate prompt = ChatPromptTemplate.from_template("tell me a short joke about {topic}") prompt_value = prompt.invoke("test_message") eas_chat_endpoint = PaiEasChatEndpoint( eas_service_url="your_service_url", eas_service_token="your_service_token" ) eas_chat_endpoint._call(prompt_value.messages) ``` It is just example code that has a potential error, so I'll explain why it's a possible type error in the description. ### Error Message and Stack Trace (if applicable) _No response_ ### Description https://github.com/langchain-ai/langchain/blob/b66a4f48fa5656871c3e849f7e1790dfb5a4c56b/libs/community/langchain_community/chat_models/pai_eas_endpoint.py#L193-L230 In the method `PaiEasChatEndpoint._call` at line 203, `self._call_eas(request_payload)` returns `str` type because it returns response.text at line 230 (https://requests.readthedocs.io/en/latest/api/#requests.Response.text). Then, first argument of `_format_response_payload` can be `str` type at line 204, resulting in type mismatch. https://github.com/langchain-ai/langchain/blob/b66a4f48fa5656871c3e849f7e1790dfb5a4c56b/libs/community/langchain_community/chat_models/pai_eas_endpoint.py#L167-L179 Moreover, if `JSONDecodeError` occurs in this method, then attribute error occurs at line 178 (`AttributeError: 'str' object has no attribute 'decode'`) because `output` variable is str type. I think `PaiEasChatEndpoint._call_eas` should be fixed to return bytes type. If I'm mistaken, I'd appreciate it if you could let me know. ### System Info System Information ------------------ > OS: Linux > OS Version: #190-Ubuntu SMP Fri Feb 2 23:24:22 UTC 2024 > Python Version: 3.9.18 (main, Feb 13 2024, 14:37:08) [GCC 9.4.0] Package Information ------------------- > langchain_core: 0.1.42 > langchain: 0.1.16 > langchain_community: 0.0.32 > langsmith: 0.1.47 > langchain_text_splitters: 0.0.1 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve

`PaiEasChatEndpoint._call_eas` should return `bytes` type instead of `str` type

https://api.github.com/repos/langchain-ai/langchain/issues/20453/comments

2024-04-15T07:47:06Z

2024-07-22T16:08:47Z

https://github.com/langchain-ai/langchain/issues/20453

### Checklist - [X] I added a very descriptive title to this issue. - [X] I included a link to the documentation page I am referring to (if applicable). ### Issue with current documentation: I read in this langchain document page: https://python.langchain.com/docs/modules/data_connection/retrievers/multi_vector/ which mentioned that can embedding summary, hypothetical question along with document. ``` The methods to create multiple vectors per document include: Smaller chunks: split a document into smaller chunks, and embed those (this is ParentDocumentRetriever). Summary: create a summary for each document, embed that along with (or instead of) the document. Hypothetical questions: create hypothetical questions that each document would be appropriate to answer, embed those along with (or instead of) the document. ``` How can I do that? In the tutorial, they just use multivector embed insteaad of the document. ``` # The vectorstore to use to index the child chunks vectorstore = Chroma(collection_name="summaries", embedding_function=OpenAIEmbeddings()) # The storage layer for the parent documents store = InMemoryByteStore() id_key = "doc_id" # The retriever (empty to start) retriever = MultiVectorRetriever( vectorstore=vectorstore, byte_store=store, id_key=id_key, ) doc_ids = [str(uuid.uuid4()) for _ in docs] summary_docs = [ Document(page_content=s, metadata={id_key: doc_ids[i]}) for i, s in enumerate(summaries) ] retriever.vectorstore.add_documents(summary_docs) retriever.docstore.mset(list(zip(doc_ids, docs))) ``` ### Idea or request for content: _No response_

DOC: MultiVector Retriever along with document embedding

https://api.github.com/repos/langchain-ai/langchain/issues/20452/comments

2024-04-15T07:28:53Z

2024-07-22T16:08:41Z

https://github.com/langchain-ai/langchain/issues/20452

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python import os os.environ["TEST_PGVECTOR_PORT"] = "5432" os.environ["TEST_PGVECTOR_DATABASE"] = "langchain_tests" os.environ["TEST_PGVECTOR_USER"] = "postgres" os.environ["TEST_PGVECTOR_PASSWORD"] = "postgres" import os from typing import List from langchain_core.embeddings import Embeddings from langchain_community.vectorstores.pgvector import PGVector class FakeEmbeddings(Embeddings): def embed_documents(self, texts: List[str]) -> List[List[float]]: return [[float(1.0)] * 9 + [float(i)] for i in range(len(texts))] def embed_query(self, text: str) -> List[float]: return [float(1.0)] * 9 + [float(0.0)] CONNECTION_STRING = PGVector.connection_string_from_db_params( driver=os.environ.get("TEST_PGVECTOR_DRIVER", "psycopg2"), host=os.environ.get("TEST_PGVECTOR_HOST", "localhost"), port=int(os.environ.get("TEST_PGVECTOR_PORT", "6024")), database=os.environ.get("TEST_PGVECTOR_DATABASE", "langchain"), user=os.environ.get("TEST_PGVECTOR_USER", "langchain"), password=os.environ.get("TEST_PGVECTOR_PASSWORD", "langchain"), ) texts = ["foo", "bar", "baz"] metadatas = [{"page": str(i)} for i in range(len(texts))] docsearch = PGVector.from_texts( texts=texts, collection_name="test_collection_filter", embedding=FakeEmbeddings(), metadatas=metadatas, connection_string=CONNECTION_STRING, pre_delete_collection=True, ) output = docsearch.similarity_search_with_score( "foo", k=3, filter={"page": {"OR": [{"EQ": "0"}, {"EQ": "2"}]}} ) # Alternatively: output = docsearch.similarity_search_with_score( "foo", k=3, filter={"page": {"AND": [{"IN": ["0","1"]}, {"NIN": ["1"]}]}} ) ``` This raises: ``` TypeError: PGVector._create_filter_clause() takes 2 positional arguments but 3 were given ``` ### Error Message and Stack Trace (if applicable) ``` Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/Users/mferenaz/Projects/personal/langchain/libs/community/langchain_community/vectorstores/pgvector.py", line 601, in similarity_search_with_score docs = self.similarity_search_with_score_by_vector( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/mferenaz/Projects/personal/langchain/libs/community/langchain_community/vectorstores/pgvector.py", line 626, in similarity_search_with_score_by_vector results = self.__query_collection(embedding=embedding, k=k, filter=filter) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/mferenaz/Projects/personal/langchain/libs/community/langchain_community/vectorstores/pgvector.py", line 945, in __query_collection filter_clauses = self._create_filter_clause_json_deprecated(filter) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/mferenaz/Projects/personal/langchain/libs/community/langchain_community/vectorstores/pgvector.py", line 828, in _create_filter_clause_json_deprecated filter_by_metadata = self._create_filter_clause_deprecated(key, value) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/mferenaz/Projects/personal/langchain/libs/community/langchain_community/vectorstores/pgvector.py", line 797, in _create_filter_clause_deprecated or_clauses = [ ^ File "/Users/mferenaz/Projects/personal/langchain/libs/community/langchain_community/vectorstores/pgvector.py", line 798, in <listcomp> self._create_filter_clause(key, sub_value) TypeError: PGVector._create_filter_clause() takes 2 positional arguments but 3 were given ``` ### Description While using PGVector without the newer jsonb format one can filter using the _create_filter_clause_deprecated method. That allows filtering with EQ, NIN, IN, etc. but also with OR and AND, the problem with this two is that the code should be calling itself but it calls the newer _create_filter_clause that does not support the same method signature and raises an error. Expected Result: When the PGVector is created without the usejsonb flag in True, should allow to filter with OR and AND conditions Actual Result: Adding an OR or AND filter without the usejsonb flag ends in an error raising ### System Info langchain==0.1.14 langchain-community==0.0.31 langchain-core==0.1.40 langchain-openai==0.0.8 langchain-text-splitters==0.0.1

PGVector deprecated create_filter_clause uses incorrect method. Can't filter by OR or AND conditions

https://api.github.com/repos/langchain-ai/langchain/issues/20445/comments

2024-04-14T22:26:26Z

2024-07-21T16:07:15Z

https://github.com/langchain-ai/langchain/issues/20445

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ``` from langchain.prompts import PromptTemplate import google.generativeai as genai from langchain_google_genai import ChatGoogleGenerativeAI from langchain_google_genai import GoogleGenerativeAI GOOGLE_API_KEY='My Api Key' llm = ChatGoogleGenerativeAI(model="gemini-pro", temperature=0, google_api_key=GOOGLE_API_KEY) prompt = "Where is Tajmahal" result = llm.invoke(prompt) ``` ### Error Message and Stack Trace (if applicable) `Retrying langchain_google_genai.chat_models._chat_with_retry.<locals>._chat_with_retry in 2.0 seconds as it raised DeadlineExceeded: 504 Deadline Exceeded.` ### Description I'm attempting to use ChatGoogleGenerativeAI, but it's running very slowly on my local Apple Silicon machine. The same code works well in Google Colab, but I'm receiving a 504 error locally. ### System Info aiohttp==3.9.4 aiosignal==1.3.1 annotated-types==0.6.0 attrs==23.2.0 cachetools==5.3.3 certifi==2024.2.2 charset-normalizer==3.3.2 dataclasses-json==0.6.4 frozenlist==1.4.1 google-ai-generativelanguage==0.6.1 google-api-core==2.18.0 google-api-python-client==2.125.0 google-auth==2.29.0 google-auth-httplib2==0.2.0 google-generativeai==0.5.0 googleapis-common-protos==1.63.0 grpcio==1.62.1 grpcio-status==1.62.1 httplib2==0.22.0 idna==3.7 jsonpatch==1.33 jsonpointer==2.4 langchain==0.1.16 langchain-community==0.0.32 langchain-core==0.1.42 langchain-google-genai==1.0.2 langchain-text-splitters==0.0.1 langsmith==0.1.47 marshmallow==3.21.1 multidict==6.0.5 mypy-extensions==1.0.0 numpy==1.26.4 orjson==3.10.0 packaging==23.2 proto-plus==1.23.0 protobuf==4.25.3 pyasn1==0.6.0 pyasn1_modules==0.4.0 pydantic==2.7.0 pydantic_core==2.18.1 pyparsing==3.1.2 PyYAML==6.0.1 requests==2.31.0 rsa==4.9 SQLAlchemy==2.0.29 tenacity==8.2.3 tqdm==4.66.2 typing-inspect==0.9.0 typing_extensions==4.11.0 uritemplate==4.1.1 urllib3==2.2.1 yarl==1.9.4 platform mac python version 3.12.3

ChatGoogleGenerativeAI is running slowly on my Apple Silicon machine.

https://api.github.com/repos/langchain-ai/langchain/issues/20444/comments

2024-04-14T21:08:53Z

2024-08-02T11:18:19Z

https://github.com/langchain-ai/langchain/issues/20444

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code The following code: ``` weaviate.Client(url=url, auth=auth, **kwargs)``` is a v3 type of implementation of Weaviate client ### Error Message and Stack Trace (if applicable) ```bash DeprecationWarning: Dep016: You are using the Weaviate v3 client, which is deprecated. Consider upgrading to the new and improved v4 client instead! See here for usage: https://weaviate.io/developers/weaviate/client-libraries/python warnings.warn(``` ### Description - I am trying to create a weaviate client using langhchain's `from langchain_community.vectorstores.weaviate import Weaviate` library. - I expected it to create a weaviate client - The current langchain internal implementaion of creating of weaveate client is working but is `v3` client API implementation which results in the warning, it should be upgraded to `v4` client API. ### System Info `pip freeze | grep langchain` : ```bash langchain==0.1.12 langchain-community==0.0.28 langchain-core==0.1.42 langchain-openai==0.0.8 langchain-pinecone==0.0.3 langchain-text-splitters==0.0.1 langchain-weaviate==0.0.1.post1 ``` - Platform: `macOS 14.4.1` - Python version: `Python 3.11.4` --------------- `python -m langchain_core.sys_info`: ```bash System Information ------------------ > OS: Darwin > OS Version: Darwin Kernel Version 23.4.0: Fri Mar 15 00:12:41 PDT 2024; root:xnu-10063.101.17~1/RELEASE_ARM64_T8103 > Python Version: 3.11.4 (main, Jul 5 2023, 08:54:11) [Clang 14.0.6 ] Package Information ------------------- > langchain_core: 0.1.42 > langchain: 0.1.16 > langchain_community: 0.0.32 > langsmith: 0.1.38 > langchain_experimental: 0.0.57 > langchain_openai: 0.1.3 > langchain_text_splitters: 0.0.1 ```

Depreciated initialisation (v3) of Weaviate Vector Database client

https://api.github.com/repos/langchain-ai/langchain/issues/20442/comments

2024-04-14T19:52:28Z

2024-05-30T10:04:52Z

https://github.com/langchain-ai/langchain/issues/20442

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code Getting error for this code - ```python from langchain_openai import OpenAI llm = OpenAI(model='gpt-3.5-turbo', temperature=0, streaming=True) llm('how are you?') ``` ### Error Message and Stack Trace (if applicable) NotFoundError: Error code: 404 - {'error': {'message': 'This is a chat model and not supported in the v1/completions endpoint. Did you mean to use v1/chat/completions?', 'type': 'invalid_request_error', 'param': 'model', 'code': None}} ### Description Getting above specified error when configuring `gpt-3.5-turbo` model with `OpenAI` However, this model works as expected with `ChatOpenAI` On the other hand, `gpt-3.5-turbo-instruct` model also works as expected with `OpenAI`, code is mentioned below - ```python from langchain_openai import OpenAI llm = OpenAI(model='gpt-3.5-turbo-instruct',temperature=0, streaming=True) llm('how are you?') ``` Here is the screenshot for reference - <img width="1193" alt="Screenshot 2024-04-15 at 12 02 51 AM" src="https://github.com/langchain-ai/langchain/assets/47742503/50cbe4de-9b94-4569-85d3-f92137aacbe4"> I believe this issue is due to configuring non-supported model with `OpenAI` instead of `ChatOpenAI` **Observation 🔍** I referred the codebase of [openai python package](https://github.com/openai/openai-python) and observed that there are some set of models which has only support of `v1/chat/completions` (`ChatOpenAI` as implemented within langchain). Check these files for more details - - `v1/chat/completions`: [src/openai/resources/chat/completions.py](https://github.com/openai/openai-python/blob/595e6b834fe26e08e9ac5415ab15bc19b8661f2c/src/openai/resources/chat/completions.py#L46) (list of specific models mentioned here) - `v1/completions`: [src/openai/resources/completions.py](https://github.com/openai/openai-python/blob/595e6b834fe26e08e9ac5415ab15bc19b8661f2c/src/openai/resources/completions.py#L39) **Potential Fix 🤔** Should we validate the model name by referring to the same list when handling parameters for OpenAI, and raise an error accordingly? **I can work on this, please check and let me know** ### System Info System Information ------------------ > OS: Darwin > OS Version: Darwin Kernel Version 23.4.0: Fri Mar 15 00:11:05 PDT 2024; root:xnu-10063.101.17~1/RELEASE_X86_64 > Python Version: 3.11.8 (main, Feb 26 2024, 15:43:17) [Clang 14.0.6 ] Package Information ------------------- > langchain_core: 0.1.42 > langchain: 0.1.16 > langchain_community: 0.0.32 > langsmith: 0.1.47 > langchain_openai: 0.1.3 > langchain_text_splitters: 0.0.1 > langchainhub: 0.1.15 > openai: 1.17.0 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve

NotFoundError for OpenAI with gpt-3.5-turbo model

https://api.github.com/repos/langchain-ai/langchain/issues/20441/comments

2024-04-14T19:07:09Z

2024-04-17T04:30:53Z

https://github.com/langchain-ai/langchain/issues/20441

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```import os from dotenv import load_dotenv from langchain_community.chat_models import ChatOllama from langchain_community.tools import DuckDuckGoSearchRun from langchain_community.utilities import DuckDuckGoSearchAPIWrapper from transformers import AutoTokenizer from langchain.tools import Tool load_dotenv() hf_token = os.environ.get("HF_TOKEN") llm = ChatOllama(model="qwen:14b") tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen1.5-14B", token=hf_token) search_wrapper = DuckDuckGoSearchAPIWrapper(region="en-us", max_results=5) # Set up the DuckDuckGo search tool search_tool = DuckDuckGoSearchRun(verbose=True, api_wrapper=search_wrapper) # Set up the search tools list search_tools = [ Tool( name="Search", func=search_tool.run, description="Use the DuckDuckGo search engine to find information", ), ] llm_with_tools = llm.bind_tools(tools=search_tools) ``` ### Error Message and Stack Trace (if applicable) ``` AttributeError Traceback (most recent call last) Cell In[1], [line 26](vscode-notebook-cell:?execution_count=1&line=26) [17](vscode-notebook-cell:?execution_count=1&line=17) # Set up the search tools list [18](vscode-notebook-cell:?execution_count=1&line=18) search_tools = [ [19](vscode-notebook-cell:?execution_count=1&line=19) Tool( [20](vscode-notebook-cell:?execution_count=1&line=20) name="Search", (...) [23](vscode-notebook-cell:?execution_count=1&line=23) ), [24](vscode-notebook-cell:?execution_count=1&line=24) ] ---> [26](vscode-notebook-cell:?execution_count=1&line=26) llm_with_tools = llm.bind_tools(tools=search_tools) AttributeError: 'ChatOllama' object has no attribute 'bind_tools' ``` ### Description In the docs for ChatOllama it says there is the method bind_tools() [here](https://api.python.langchain.com/en/latest/chat_models/langchain_community.chat_models.ollama.ChatOllama.html#langchain_community.chat_models.ollama.ChatOllama.bind_tools). When I attempt to run the Ollama instance and bind the search tool it throws the AttributeError saying it has no attribute 'bind_tools' ### System Info System Information ------------------ > OS: Darwin > OS Version: Darwin Kernel Version 23.4.0: Fri Mar 15 00:10:42 PDT 2024; root:xnu-10063.101.17~1/RELEASE_ARM64_T6000 > Python Version: 3.12.2 | packaged by conda-forge | (main, Feb 16 2024, 20:54:21) [Clang 16.0.6 ] Package Information ------------------- > langchain_core: 0.1.40 > langchain: 0.1.14 > langchain_community: 0.0.31 > langsmith: 0.1.38 > langchain_experimental: 0.0.56 > langchain_text_splitters: 0.0.1 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve

'ChatOllama' object has no attribute 'bind_tools'

https://api.github.com/repos/langchain-ai/langchain/issues/20439/comments

2024-04-14T18:08:10Z

2024-08-04T08:48:34Z

https://github.com/langchain-ai/langchain/issues/20439

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ``` client = AsyncOpenAI( # Defaults to os.environ.get("OPENAI_API_KEY") # Otherwise use: api_key="Your_API_Key", api_key=settings.OPENAI_API_KEY, ) async def acreate_assistant(**kwargs: Any, ): return await OpenAIAssistantRunnable.acreate_assistant(async_client=client, **kwargs) ``` when I run `await acreate_assistant()` it shows `Unexpected error occurred: OpenAIError - The api_key client option must be set either by passing api_key to the client or by setting the OPENAI_API_KEY environment variable` ### Error Message and Stack Trace (if applicable) _No response_ ### Description Here is the code of `client` property in `OpenAIAssistantRunnable` ``` client: Any = Field(default_factory=_get_openai_client) ``` Because of using `AsyncClient`, the `client` value is None as default. So it will create by `_get_openai_client`. ``` class OpenAI(SyncAPIClient): def __init__(...): if api_key is None: api_key = os.environ.get("OPENAI_API_KEY") if api_key is None: raise OpenAIError( "The api_key client option must be set either by passing api_key to the client or by setting the OPENAI_API_KEY environment variable" ) self.api_key = api_key ``` but the `OpenAI` will check the `api_key` automaticlly. So it will always failed when I use AsyncOpenAIClient to create the `OpenAIAssistantRunnable` object. I don't know if it's Ok to remove the `default_factory` of `client`. If it's ok, I'll create a PR to fix this problem. ### System Info ``` langchain==0.1.16 langchain-community==0.0.32 langchain-core==0.1.42 langchain-openai==0.1.0 langchain-text-splitters==0.0.1 langsmith==0.1.31 ```

create openai assistant object failed when I use AsyncOpenAI client

https://api.github.com/repos/langchain-ai/langchain/issues/20436/comments

2024-04-14T15:03:37Z

2024-08-05T16:08:41Z

https://github.com/langchain-ai/langchain/issues/20436

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python from langchain_community.embeddings import LocalAIEmbeddings from langchain_community.vectorstores import FAISS embeddings_model = LocalAIEmbeddings( openai_api_base=OPENAI_BASE_URL, openai_api_key=OPENAI_API_KEY, model="bge-large-zh-v1.5" ) db = FAISS.load_local("data.db", embeddings_model, allow_dangerous_deserialization=True) ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description AttributeError: module ‘openai’ has no attribute ‘error’ ### System Info langchain 0.1.16 linux Python :: 3.11

How to specify custom embeddings?

https://api.github.com/repos/langchain-ai/langchain/issues/20428/comments

2024-04-14T01:55:33Z

2024-07-21T16:07:05Z

https://github.com/langchain-ai/langchain/issues/20428

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code xccxcx ### Error Message and Stack Trace (if applicable) MATCH (n:`category`) WHERE n.embedding IS null AND any(k in $props WHERE n[k] IS NOT null) RETURN elementId(n) AS id, reduce(str='',k IN $props | str + '\\n' + k + ':' + coalesce(n[k], '')) AS text LIMIT 1000 ### Description The query residing at line 1278 always add propetry name, semicolon and new line to all queried data so distorts embeddings. MATCH (n:`category`) WHERE n.embedding IS null AND any(k in $props WHERE n[k] IS NOT null) RETURN elementId(n) AS id, reduce(str='',k IN $props | str + '\\n' + k + ':' + coalesce(n[k], '')) AS text LIMIT 1000 I am suggesting to remove this part of the query " '\\n' + k + ':' " ### System Info The bug is multi platform.

Method from_existing_graph() distorts queried data from Neo4J so impacts embeddings.

https://api.github.com/repos/langchain-ai/langchain/issues/20423/comments

2024-04-13T23:29:05Z

2024-07-20T16:07:25Z

https://github.com/langchain-ai/langchain/issues/20423

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ** ### Error Message and Stack Trace (if applicable) ** ### Description The Voyage AI Reranker cannot be imported as a module without #19645 ### System Info **

Request for a langchain-voyageai package release

https://api.github.com/repos/langchain-ai/langchain/issues/20408/comments

2024-04-13T07:22:51Z

2024-07-30T16:06:45Z

https://github.com/langchain-ai/langchain/issues/20408

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [ ] I am sure that this is a bug in LangChain rather than my code. - [ ] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code Hi I am trying to generate QA from a pdf Running following code snippet - text_splitter = RecursiveCharacterTextSplitter( # Set a really small chunk size, just to show. chunk_size = 1000, chunk_overlap = 100, length_function = len, ) loader = PyPDFLoader(data_pdf) pages = loader.load_and_split(text_splitter=text_splitter) len(pages) #260 templ = """You are a smart assistant designed to come up with meaninful question and answer pair. The question should be to the point and the answer should be as detailed as possible. Given a piece of text, you must come up with a question and answer pair that can be used to evaluate a QA bot. Do not make up stuff. Stick to the text to come up with the question and answer pair. When coming up with this question/answer pair, you must respond in the following format: ``` {{ "question": "$YOUR_QUESTION_HERE", "answer": "$THE_ANSWER_HERE" }} ``` Everything between the ``` must be valid json. Please come up with a question/answer pair, in the specified JSON format, for the following text: ---------------- {text}""" PROMPT = PromptTemplate.from_template(templ) llm = Cohere(model="command", temperature=0) # command, command-light chain = QAGenerationChain.from_llm(llm=llm, prompt=PROMPT) ### llm is as follows ##Cohere(client=<cohere.client.Client object at 0x00000188268D3BD0>, async_client=<cohere.client.AsyncClient object at ##0x0000018827B22690>, model='command', temperature=0.0) #By running following code I am expecting QA set to be generated but I am getting error chain.invoke(pages[40].page_content) I tried other models like OpenAI and Google Gemini Pro and QAGeneration chain fails at same step ### Error Message and Stack Trace (if applicable) --------------------------------------------------------------------------- AttributeError Traceback (most recent call last) Cell In[17], line 1 ----> 1 chain.invoke(pages[40].page_content) File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain\chains\base.py:163, in Chain.invoke(self, input, config, **kwargs) 161 except BaseException as e: 162 run_manager.on_chain_error(e) --> 163 raise e 164 run_manager.on_chain_end(outputs) 166 if include_run_info: File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain\chains\base.py:153, in Chain.invoke(self, input, config, **kwargs) 150 try: 151 self._validate_inputs(inputs) 152 outputs = ( --> 153 self._call(inputs, run_manager=run_manager) 154 if new_arg_supported 155 else self._call(inputs) 156 ) 158 final_outputs: Dict[str, Any] = self.prep_outputs( 159 inputs, outputs, return_only_outputs 160 ) 161 except BaseException as e: File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain\chains\qa_generation\base.py:73, in QAGenerationChain._call(self, inputs, run_manager) 67 def _call( 68 self, 69 inputs: Dict[str, Any], 70 run_manager: Optional[CallbackManagerForChainRun] = None, 71 ) -> Dict[str, List]: 72 docs = self.text_splitter.create_documents([inputs[self.input_key]]) ---> 73 results = self.llm_chain.generate( 74 [{"text": d.page_content} for d in docs], run_manager=run_manager 75 ) 76 qa = [json.loads(res[0].text) for res in results.generations] 77 return {self.output_key: qa} File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain\chains\llm.py:115, in LLMChain.generate(self, input_list, run_manager) 113 callbacks = run_manager.get_child() if run_manager else None 114 if isinstance(self.llm, BaseLanguageModel): --> 115 return self.llm.generate_prompt( 116 prompts, 117 stop, 118 callbacks=callbacks, 119 **self.llm_kwargs, 120 ) 121 else: 122 results = self.llm.bind(stop=stop, **self.llm_kwargs).batch( 123 cast(List, prompts), {"callbacks": callbacks} 124 ) File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain_core\language_models\llms.py:597, in BaseLLM.generate_prompt(self, prompts, stop, callbacks, **kwargs) 589 def generate_prompt( 590 self, 591 prompts: List[PromptValue], (...) 594 **kwargs: Any, 595 ) -> LLMResult: 596 prompt_strings = [p.to_string() for p in prompts] --> 597 return self.generate(prompt_strings, stop=stop, callbacks=callbacks, **kwargs) File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain_core\language_models\llms.py:767, in BaseLLM.generate(self, prompts, stop, callbacks, tags, metadata, run_name, run_id, **kwargs) 752 if (self.cache is None and get_llm_cache() is None) or self.cache is False: 753 run_managers = [ 754 callback_manager.on_llm_start( 755 dumpd(self), (...) 765 ) 766 ] --> 767 output = self._generate_helper( 768 prompts, stop, run_managers, bool(new_arg_supported), **kwargs 769 ) 770 return output 771 if len(missing_prompts) > 0: File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain_core\language_models\llms.py:634, in BaseLLM._generate_helper(self, prompts, stop, run_managers, new_arg_supported, **kwargs) 632 for run_manager in run_managers: 633 run_manager.on_llm_error(e, response=LLMResult(generations=[])) --> 634 raise e 635 flattened_outputs = output.flatten() 636 for manager, flattened_output in zip(run_managers, flattened_outputs): File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain_core\language_models\llms.py:621, in BaseLLM._generate_helper(self, prompts, stop, run_managers, new_arg_supported, **kwargs) 611 def _generate_helper( 612 self, 613 prompts: List[str], (...) 617 **kwargs: Any, 618 ) -> LLMResult: 619 try: 620 output = ( --> 621 self._generate( 622 prompts, 623 stop=stop, 624 # TODO: support multiple run managers 625 run_manager=run_managers[0] if run_managers else None, 626 **kwargs, 627 ) 628 if new_arg_supported 629 else self._generate(prompts, stop=stop) 630 ) 631 except BaseException as e: 632 for run_manager in run_managers: File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain_core\language_models\llms.py:1231, in LLM._generate(self, prompts, stop, run_manager, **kwargs) 1228 new_arg_supported = inspect.signature(self._call).parameters.get("run_manager") 1229 for prompt in prompts: 1230 text = ( -> 1231 self._call(prompt, stop=stop, run_manager=run_manager, **kwargs) 1232 if new_arg_supported 1233 else self._call(prompt, stop=stop, **kwargs) 1234 ) 1235 generations.append([Generation(text=text)]) 1236 return LLMResult(generations=generations) File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain_community\llms\cohere.py:217, in Cohere._call(self, prompt, stop, run_manager, **kwargs) 202 """Call out to Cohere's generate endpoint. 203 204 Args: (...) 214 response = cohere("Tell me a joke.") 215 """ 216 params = self._invocation_params(stop, **kwargs) --> 217 response = completion_with_retry( 218 self, model=self.model, prompt=prompt, **params 219 ) 220 _stop = params.get("stop_sequences") 221 return self._process_response(response, _stop) File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain_community\llms\cohere.py:45, in completion_with_retry(llm, **kwargs) 43 def completion_with_retry(llm: Cohere, **kwargs: Any) -> Any: 44 """Use tenacity to retry the completion call.""" ---> 45 retry_decorator = _create_retry_decorator(llm) 47 @retry_decorator 48 def _completion_with_retry(**kwargs: Any) -> Any: 49 return llm.client.generate(**kwargs) File ~\anaconda3\envs\GenAI_py311\Lib\site-packages\langchain_community\llms\cohere.py:38, in _create_retry_decorator(llm) 31 max_seconds = 10 32 # Wait 2^x * 1 second between each retry starting with 33 # 4 seconds, then up to 10 seconds, then 10 seconds afterwards 34 return retry( 35 reraise=True, 36 stop=stop_after_attempt(llm.max_retries), 37 wait=wait_exponential(multiplier=1, min=min_seconds, max=max_seconds), ---> 38 retry=(retry_if_exception_type(cohere.error.CohereError)), 39 before_sleep=before_sleep_log(logger, logging.WARNING), 40 ) AttributeError: module 'cohere' has no attribute 'error' ### Description chain = QAGenerationChain.from_llm(llm=llm, prompt=PROMPT) chain.invoke(pages[40].page_content) Chain.run and chain.invoke giving error ### System Info python 3.11 langchain==0.1.9 langchain-community==0.0.24 langchain-core==0.1.42 langchain-google-genai==1.0.2

Error run or invoke method of QAGenerationChain

https://api.github.com/repos/langchain-ai/langchain/issues/20406/comments

2024-04-13T04:48:27Z

2024-07-21T16:07:00Z

https://github.com/langchain-ai/langchain/issues/20406

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code I have the following code: ```python from langchain_community.agent_toolkits.sql.base import create_sql_agent from langchain_community.utilities.sql_database import SQLDatabase db = SQLDatabase.from_uri(database_uri="%POSTGRES_URI%") prompt = "what is the total corpus of money in banks" conversation = create_sql_agent(llm=llm, db=db, agent_type="openai-tools", verbose=True, top_k=15) return conversation.invoke(input={"input": prompt}) ``` ### Error Message and Stack Trace (if applicable) > Entering new SQL Agent Executor chain... INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 200 OK" Invoking: `sql_db_list_tables` with `{}` ERROR: 1 validation error for _ListSQLDataBaseToolInput tool_input field required (type=value_error.missing) ### Description I recently updated langchain and pydantic to their latest version and my code broke and when using the sql agent I get input missing issue. I think the invoking of sql_db_list_tables is not putting any value into when trying to parse the description of table schema. And this is a bug from langchain library itself. Kindly help into this issue, as this is becoming a blocker. Thanks. ### System Info System Information ------------------ > OS: Darwin > OS Version: Darwin Kernel Version 23.1.0: Mon Oct 9 21:27:24 PDT 2023; root:xnu-10002.41.9~6/RELEASE_ARM64_T6000 > Python Version: 3.11.7 (main, Dec 4 2023, 18:10:11) [Clang 15.0.0 (clang-1500.1.0.2.5)] Package Information ------------------- > langchain_core: 0.1.42 > langchain: 0.1.16 > langchain_community: 0.0.32 > langsmith: 0.1.45 > langchain_experimental: 0.0.50 > langchain_openai: 0.1.3 > langchain_pinecone: 0.0.3 > langchain_text_splitters: 0.0.1 > langchainhub: 0.1.14 > pydantic-settings: 2.1.0 > pydantic: 2.7.0 > pydantic-core: 2.18.1 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve

[BUG] Getting validation error for SQL Database, type=value_error.missing, when its trying to list tables in the db

https://api.github.com/repos/langchain-ai/langchain/issues/20405/comments

2024-04-13T04:14:38Z

2024-08-07T16:06:20Z

https://github.com/langchain-ai/langchain/issues/20405

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code I am ingesting docs to qdrant using langchain but not able to retrieve a list of chunks from the vector store. My ingest.py script: ```python from langchain.text_splitter import RecursiveCharacterTextSplitter from langchain_openai import OpenAIEmbeddings from langchain_community.document_loaders import PDFMinerLoader from langchain_community.vectorstores.qdrant import Qdrant from dotenv import load_dotenv import uuid load_dotenv() OPENAI_API_TOKEN = os.getenv('OPENAI_API_KEY') def load_documents(): texts = [] all_items = os.listdir("files/") for _, item in enumerate(all_items): print("Processing file: {}".format(item)) unique_id = str(uuid.uuid4()) text_splitter = RecursiveCharacterTextSplitter() loader = PDFMinerLoader("files/{}".format(item), extract_images=True) docs = loader.load_and_split(text_splitter) for doc in docs: doc.metadata["pdf"] = doc.metadata["source"] doc.metadata["unique_id"] = unique_id texts.extend(docs) return texts def create_qdrant_store(texts): print("Creating qdrant store") embeddings = OpenAIEmbeddings(openai_api_key=OPENAI_API_TOKEN, model="text-embedding-3-small") Qdrant.from_documents( texts, embeddings, url="http://localhost:6333", prefer_grpc=True, collection_name="aff_container", ) def main(): texts = load_documents() print(texts) create_qdrant_store(texts) print("Docuemnts loaded successfully!") if __name__ == "__main__": main() ``` My get files endpoint: ```python @app.route('/api/get_files', methods=['GET']) def get_files(): """ Endpoint to get the list of files in the index. """ try: client = QdrantClient(url="http://localhost:6333") db = Qdrant(client=client, collection_name='affilix_container', embeddings=embeddings) files = [] ## ----- want to list all the chunks here ----- ## print(client.get_collection(collection_name='affilix_container').) for doc in db.collection_name: file_id = doc.metadata['unique_id'] file_name = doc.metadata['pdf'].replace("files/", "").replace("temp/", "") if file_id not in [file['id'] for file in files]: files.append({'id': file_id, 'file_name': file_name}) return jsonify({'files': files}) except Exception as e: return jsonify({'error': str(e)}), 500` ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description I am ingesting docs to qdrant using langchain but not able to retrieve a list of chunks from the vector store. ### System Info $ pip freeze | grep langchain langchain==0.1.9 langchain-community==0.0.24 langchain-core==0.1.40 langchain-openai==0.0.8

Not able to list all the documents from qdrant vector store using langchain

https://api.github.com/repos/langchain-ai/langchain/issues/20382/comments

2024-04-12T10:36:34Z

2024-04-13T09:30:09Z

https://github.com/langchain-ai/langchain/issues/20382

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code Definition of the output using PydanticOutputParser: ```python class Joke(BaseModel): setup: str = Field(description="question to set up a joke") punchline: str = Field(description="answer to resolve the joke") parser = PydanticOutputParser(pydantic_object=Joke) ``` ### Error Message and Stack Trace (if applicable) ``` Error in RootListenersTracer.on_chain_end callback: ValueError() ``` ### Description Since I'm trying to build a conversational agent, I am using [RunnableWithMessageHistory](https://python.langchain.com/docs/expression_language/how_to/message_history/) to automatically add the messages to the history. I would expect everything to work fine, but I obtain the following error ``` Error in RootListenersTracer.on_chain_end callback: ValueError() ``` from ```python def _get_output_messages( self, output_val: Union[str, BaseMessage, Sequence[BaseMessage], dict] ) -> List[BaseMessage]: from langchain_core.messages import BaseMessage if isinstance(output_val, dict): output_val = output_val[self.output_messages_key or "output"] if isinstance(output_val, str): from langchain_core.messages import AIMessage return [AIMessage(content=output_val)] elif isinstance(output_val, BaseMessage): return [output_val] elif isinstance(output_val, (list, tuple)): return list(output_val) else: raise ValueError() ``` Since I used PydanticOutputParser, here `output_val `is a type dict object containing a Joke object: ``` output_val = {'output': Joke(joke='Why did the egg refuse to tell jokes? Because it cracked under pressure!', message='Here a joke for you!')} ``` Joke objects do not seem to be supported by `_get_output_messages`. How can I get it to work? I would like for the history to update automatically when using RunnableWithMessageHistory and a PydanticOutputParser. ### System Info langchain==0.1.16 langchain-cli==0.0.19 langchain-community==0.0.32 langchain-core==0.1.42 langchain-experimental==0.0.10 langchain-openai==0.0.8 langchain-text-splitters==0.0.1 python version 3.10.9 OS: windows11

How to use RunnableWithMessageHistory?

https://api.github.com/repos/langchain-ai/langchain/issues/20380/comments

2024-04-12T10:18:41Z

2024-04-16T13:08:19Z

https://github.com/langchain-ai/langchain/issues/20380

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code from langchain_community.tools import BearlyInterpreterTool bearly_tool.add_file( source_path="sample_data/US_GDP.csv", target_path="US_GDP.csv", description="" ) agent = initialize_agent( tools, llm, agent=AgentType.OPENAI_FUNCTIONS, verbose=True, handle_parsing_errors=True, ) # Simple Queries agent.run("What was the US GDP in 2019?") raise self._make_status_error_from_response(err.response) from None openai.BadRequestError: Error code: 400 - {'error': {'message': '\'Evaluates python code in a sandbox environment. The environment resets on every execution. You must send the whole script every time and print your outputs. Script should be pure python code that can be evaluated. It should be in python format NOT markdown. The code should NOT be wrapped in backticks. All python packages including requests, matplotlib, scipy, numpy, pandas, etc are available. If you have any files outputted write them to "output/" relative to the execution path. Output can only be read from the directory, stdout, and stdin. Do not use things like plot.show() as it will not work instead write them out `output/` and a link to the file will be returned. print() any output and results so you can capture the output.\\n\\nThe following files available in the evaluation environment: ### Error Message and Stack Trace (if applicable) _No response_ ### Description i want to use BearlyInterpreterTool,but get erro ### System Info python version3.10 liunx langchain 0.1.14

openai.BadRequestError: Error code: 400 - {'error': {'message': '\'Evaluates python code in a sandbox environment The environment resets on every execution. You must send the whole script every time and print your outputs. Script should be pure python code that can be evaluated. It should be in python format NOT markdown.

https://api.github.com/repos/langchain-ai/langchain/issues/20377/comments

2024-04-12T09:25:08Z

2024-07-19T16:08:30Z

https://github.com/langchain-ai/langchain/issues/20377

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python from typing import Any, AsyncIterator, List, Sequence, cast from langchain_core.runnables.schema import StreamEvent from langchain.agents import AgentExecutor, create_openai_tools_agent from langchain_openai import ChatOpenAI from langchain_core.prompts.chat import ( ChatPromptTemplate, SystemMessagePromptTemplate, HumanMessagePromptTemplate, MessagesPlaceholder, PromptTemplate, ) import langchain_core import typing from langchain_core.documents import Document from langchain_core.tools import tool def foo(x: int) -> dict: """Foo""" return {"x": 5} @tool def get_docs(x: int) -> list[Document]: """get_docs""" return [Document(page_content="hello")] def _with_nulled_run_id(events: Sequence[StreamEvent]) -> List[StreamEvent]: """Removes the run ids from events.""" return cast(List[StreamEvent], [{**event, "run_id": ""} for event in events]) async def _collect_events(events: AsyncIterator[StreamEvent]) -> List[StreamEvent]: """Collect the events and remove the run ids.""" materialized_events = [event async for event in events] events_ = _with_nulled_run_id(materialized_events) for event in events_: event["tags"] = sorted(event["tags"]) return events_ prompt_obj = { "name": None, "input_variables": ["agent_scratchpad", "input"], "input_types": { "chat_history": typing.List[ typing.Union[ langchain_core.messages.ai.AIMessage, langchain_core.messages.human.HumanMessage, langchain_core.messages.chat.ChatMessage, langchain_core.messages.system.SystemMessage, langchain_core.messages.function.FunctionMessage, langchain_core.messages.tool.ToolMessage, ] ], "agent_scratchpad": typing.List[ typing.Union[ langchain_core.messages.ai.AIMessage, langchain_core.messages.human.HumanMessage, langchain_core.messages.chat.ChatMessage, langchain_core.messages.system.SystemMessage, langchain_core.messages.function.FunctionMessage, langchain_core.messages.tool.ToolMessage, ] ], }, "output_parser": None, "partial_variables": {}, "metadata": { "lc_hub_owner": "hwchase17", "lc_hub_repo": "openai-tools-agent", "lc_hub_commit_hash": "c18672812789a3b9697656dd539edf0120285dcae36396d0b548ae42a4ed66f5", }, "tags": None, "messages": [ SystemMessagePromptTemplate(prompt=PromptTemplate(input_variables=[], template="You are a helpful assistant")), MessagesPlaceholder(variable_name="chat_history", optional=True), HumanMessagePromptTemplate(prompt=PromptTemplate(input_variables=["input"], template="{input}")), MessagesPlaceholder(variable_name="agent_scratchpad"), ], "validate_template": False, } prompt = ChatPromptTemplate.parse_obj(prompt_obj) tools = [get_docs] llm = ChatOpenAI(model="gpt-3.5-turbo-1106", temperature=0) # Construct the OpenAI Tools agent agent = create_openai_tools_agent(llm, tools, prompt) # Create an agent executor by passing in the agent and tools agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True) events = await _collect_events( agent_executor.astream_events({"input": "call get_docs."}, version="v1", include_names=["get_docs"]) ) assert events == [ { "event": "on_tool_start", "name": "get_docs", "run_id": "", "tags": [], "metadata": {}, "data": {"input": {"x": 5}}, }, { "event": "on_tool_end", "name": "get_docs", "run_id": "", "tags": [], "metadata": {}, "data": {"input": {"x": 5}, "output": [Document(page_content="hello")]}, }, ] ``` ### Error Message and Stack Trace (if applicable) Assertion error: ``` --------------------------------------------------------------------------- { "name": "AssertionError", "message": "", "stack": "--------------------------------------------------------------------------- AssertionError Traceback (most recent call last) Cell In[4], line 96 92 agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True) 93 events = await _collect_events( 94 agent_executor.astream_events({\"input\": \"call get_docs.\"}, version=\"v1\", include_names=[\"get_docs\"]) 95 ) ---> 96 assert events == [ 97 { 98 \"event\": \"on_tool_start\", 99 \"name\": \"get_docs\", 100 \"run_id\": \"\", 101 \"tags\": [], 102 \"metadata\": {}, 103 \"data\": {\"input\": {\"x\": 5}}, 104 }, 105 { 106 \"event\": \"on_tool_end\", 107 \"name\": \"get_docs\", 108 \"run_id\": \"\", 109 \"tags\": [], 110 \"metadata\": {}, 111 \"data\": {\"input\": {\"x\": 5}, \"output\": [Document(page_content=\"hello\")]}, 112 }, 113 ] AssertionError: " } ``` ### Description When using an agent executor, and we call a tool, I expect the actual output of the tool, rather than the output being cast to a string. This bug was originally raised [here](https://github.com/langchain-ai/langchain/discussions/18694), and partially fixed except for when using an agent executor in this [PR](https://github.com/langchain-ai/langchain/pull/18932). This comment shows the cause of the issue: https://github.com/langchain-ai/langchain/pull/18932#issuecomment-2034933719 ### System Info ``` System Information ------------------ > OS: Linux > OS Version: #1 SMP Thu Oct 5 21:02:42 UTC 2023 > Python Version: 3.10.6 | packaged by conda-forge | (main, Aug 22 2022, 20:35:26) [GCC 10.4.0] Package Information ------------------- > langchain_core: 0.1.42 > langchain: 0.1.16 > langchain_community: 0.0.32 > langsmith: 0.1.38 > langchain_experimental: 0.0.57 > langchain_openai: 0.1.3 > langchain_text_splitters: 0.0.1 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve ```

AsyncCallbackManagerForToolRun improperly casts on_tool_end to string

https://api.github.com/repos/langchain-ai/langchain/issues/20372/comments

2024-04-12T07:16:02Z

2024-07-19T16:08:26Z

https://github.com/langchain-ai/langchain/issues/20372

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code class AnalyzeResume(BaseModel): """As a professional HR agent that helps users analyze and generate a comprehensive report about the resume.\n Use the following pieces of context from a resume to answer the question at the end.""" issues: list = Field( description="List down all issues detected in the resume with **title, description, solution format** with description & solution of atleast 600 words" ) suggestions: list = Field( description="List down all suggestions to improve the resume in **title, description format** with description of atleast 600 words" ) summary: str = Field(description="Summary of the of the resume.") negotiations: list = Field( description="List down all insightful points for resume negotiations in **title, description format** with description of atleast 600 words" ) companies: list = Field( description="List down all Names and descriptions of the companies involved in the resume in **title, description format**" ) ### Error Message and Stack Trace (if applicable)  ### Description I want the output should be consistent, sometimes the output list is having data, sometimes just completely blank. Below is the prompt code I am using ``` resume_parser = PydanticOutputParser(pydantic_object=AnalyzeResume) template = """ As a professional HR agent that helps users analyze and generate a comprehensive report about the resume.\ Use the following pieces of context from a resume to answer the question at the end.\ \n {context}\n \n{format_instructions}\n Question: {question} """ rag_prompt_custom = PromptTemplate( template=template, input_variables=["context", "question"], partial_variables={"format_instructions": resume_parser.get_format_instructions()}, ) chroma_client = chromadb.HttpClient(host="chromadb", port=8000) document_analysis_client = DocumentAnalysisClient( endpoint=os.environ["AZURE_DI_SERVICE_ENDPOINT"], credential=AzureKeyCredential(os.environ["AZURE_DI_API_KEY"]), mode="page", analysis_features=["ocrHighResolution"], ) loader = DocumentIntelligenceLoader( file_path=contract_obj.file_input.path, client=document_analysis_client, model="prebuilt-document" ) documents = loader.load() text_splitter = RecursiveCharacterTextSplitter( chunk_size=14500, chunk_overlap=100, add_start_index=True, strip_whitespace=True, separators=MARKDOWN_SEPARATORS, ) docs = text_splitter.split_documents(documents) collection_id = str(uuid.uuid4()) chroma_db = Chroma.from_documents(docs, azure_embeddings, client=chroma_client, collection_name=collection_id) contract_obj.collection_id = collection_id contract_obj.save() vectorstore = Chroma(client=chroma_client, collection_name=collection_id, embedding_function=azure_embeddings) retriever = vectorstore.as_retriever() rag_chain = {"context": retriever, "question": RunnablePassthrough()} | rag_prompt_custom | azure_llm with get_openai_callback() as cb: result = rag_chain.invoke("Generate answers in provided format") total_rag_tokens = cb.total_tokens ``` Just point me to the issue, I will try my best to fix it. (Might be chroma or Azure OpenAI or the text splitter) ### System Info langchain langchain-community langchain-core langchain-google-genai==0.0.8 langchain-openai==0.0.5 langchainhub langsmith azure-ai-documentintelligence==1.0.0b1 azure-ai-formrecognizer==3.3.2 chroma-hnswlib==0.7.3 chromadb == 0.4.22

Highly inconsistent output using PydanticOutputParser

https://api.github.com/repos/langchain-ai/langchain/issues/20369/comments

2024-04-12T06:31:30Z

2024-04-12T21:20:44Z

https://github.com/langchain-ai/langchain/issues/20369

### Checked other resources - [x] I added a very descriptive title to this issue. - [x] I searched the LangChain documentation with the integrated search. - [x] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code def generate_response_langchain(user_question, complexity_level, memory): """ Generate a response based on user question using LangChain """ groq_chat = None if complexity_level == "simple": groq_chat = ChatGroq( groq_api_key model_name='llama2-70b-4096' # LLaMA2-70b model ) elif complexity_level == "large": groq_chat = ChatGroq( groq_api_key= model_name='mixtral-8x7b-32768' # Mixtral-8x7b model ) elif complexity_level == "complex": groq_chat = ChatGroq( groq_api_key= model_name='gemma-7b-it' # Gemma-7b-it model ) conversation = ConversationChain( llm=groq_chat, memory=memory ) conversational_memory_length = 100 if 'chat_history' not in st.session_state: st.session_state.chat_history = [] memory = ConversationBufferWindowMemory(k=conversational_memory_length) # Display chat history st.write("Chat History:") for sender, message in st.session_state.chat_history: st.write(f"{sender} {message}") user_question = st.chat_input("Say something") if user_question: st.session_state.chat_history.append(("User:", user_question)) ### Error Message and Stack Trace (if applicable) def generate_response_langchain(user_question, complexity_level, memory): """ Generate a response based on user question using LangChain """ groq_chat = None if complexity_level == "simple": groq_chat = ChatGroq( groq_api_key model_name='llama2-70b-4096' # LLaMA2-70b model ) elif complexity_level == "large": groq_chat = ChatGroq( groq_api_key= model_name='mixtral-8x7b-32768' # Mixtral-8x7b model ) elif complexity_level == "complex": groq_chat = ChatGroq( groq_api_key= model_name='gemma-7b-it' # Gemma-7b-it model ) conversation = ConversationChain( llm=groq_chat, memory=memory ) conversational_memory_length = 100 if 'chat_history' not in st.session_state: st.session_state.chat_history = [] memory = ConversationBufferWindowMemory(k=conversational_memory_length) # Display chat history st.write("Chat History:") for sender, message in st.session_state.chat_history: st.write(f"{sender} {message}") user_question = st.chat_input("Say something") if user_question: st.session_state.chat_history.append(("User:", user_question)) ### Description i have a problem with chat history if i told him my name in next message, if i asked about my name the model cant answer "Forget the name." i need to make history-based Contextualization cant i get your help? ### System Info 

i have a problem with chat history if i told him my name in next message, if i asked about my name the model cant answer "Forget the name." i need to make history-based Contextualization cant i get your help?

https://api.github.com/repos/langchain-ai/langchain/issues/20367/comments

2024-04-12T04:11:04Z

2024-07-19T16:08:20Z

https://github.com/langchain-ai/langchain/issues/20367

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ``` from langchain_community.llms import HuggingFaceEndpoint llm = HuggingFaceEndpoint( endpoint_url="http://localhost:8010/", max_new_tokens=512, top_k=10, top_p=0.95, typical_p=0.95, temperature=0.01, repetition_penalty=1.03, huggingfacehub_api_token=None ) ``` ### Error Message and Stack Trace (if applicable) ``` --------------------------------------------------------------------------- ValidationError Traceback (most recent call last) Cell In[2], line 3 1 from langchain_community.llms import HuggingFaceEndpoint ----> 3 llm = HuggingFaceEndpoint( 4 endpoint_url="http://localhost:8010/", 5 max_new_tokens=512, 6 top_k=10, 7 top_p=0.95, 8 typical_p=0.95, 9 temperature=0.01, 10 repetition_penalty=1.03, 11 huggingfacehub_api_token=None 12 ) File ~/Jupyter/llm/venv/lib/python3.10/site-packages/langchain_core/load/serializable.py:120, in Serializable.__init__(self, **kwargs) 119 def __init__(self, **kwargs: Any) -> None: --> 120 super().__init__(**kwargs) 121 self._lc_kwargs = kwargs File ~/Jupyter/llm/venv/lib/python3.10/site-packages/pydantic/v1/main.py:341, in BaseModel.__init__(__pydantic_self__, **data) 339 values, fields_set, validation_error = validate_model(__pydantic_self__.__class__, data) 340 if validation_error: --> 341 raise validation_error 342 try: 343 object_setattr(__pydantic_self__, '__dict__', values) ValidationError: 1 validation error for HuggingFaceEndpoint __root__ Could not authenticate with huggingface_hub. Please check your API token. (type=value_error) ``` ### Description #### Background While restructuring our codebase in response to the deprecation of `HuggingFaceTextGenInference`, I encountered an error when attempting to create a `HuggingFaceEndpoint` with a locally hosted [TGI server](https://github.com/huggingface/text-generation-inference). #### Issue The error occurs in the `validate_environment function` of the `huggingface_endpoint.py` file, specifically in the lines [170-179](https://github.com/langchain-ai/langchain/blob/master/libs/community/langchain_community/llms/huggingface_endpoint.py#L170-L179). The `@root_validator()` decorator throws an error when `huggingfacehub_api_token` is passed as `None`, which happens due to `login(token=huggingfacehub_api_token)` in `validate_environment` function. By commenting out the block that processes the API token and manually setting `huggingfacehub_api_token` to `None`, I am able to successfully create an `InferenceClient`. Since HuggingFaceTextGenInference is fused into HuggingFaceEndpoint in PR #17254, we need to add logic to handle cases where `huggingfacehub_api_token` is passed as `None` or when no environment variable `HUGGINGFACEHUB_API_TOKEN` is set. This is particularly necessary for setups using a locally hosted TGI server where authentication with the Huggingface Hub may not be required. ### System Info huggingface-hub==0.22.2 langchain-commnity==0.0.32 platform: linux python version: 3.10

Handling huggingfacehub_api_token=None for HuggingFaceEndpoint

https://api.github.com/repos/langchain-ai/langchain/issues/20342/comments

2024-04-11T17:31:23Z

2024-06-03T22:20:33Z

https://github.com/langchain-ai/langchain/issues/20342

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ``` from uuid import uuid4 from langchain.agents import AgentExecutor, create_structured_chat_agent from langchain.prompts import ( ChatPromptTemplate, HumanMessagePromptTemplate, MessagesPlaceholder, SystemMessagePromptTemplate, ) from langchain.memory import ConversationBufferWindowMemory def __handle_parsing_error(self, error): self.__logger.error(f"Parsing error encountered: {error}") # For now, returning a generic error message. return "I'm sorry, there was a problem understanding your request." def __get_agent(self, session_id): return AgentExecutor.from_agent_and_tools( agent=self.__agent, tools=self.__tools, verbose=self.__verbose_mode, memory=self.__get_history(session_id), handle_parsing_errors=__handle_parsing_error, return_intermediate_steps=False, ) def send_message(self, session_id, message: str = "") -> str: if not message.strip(): return "You didn't ask a question. How can I assist you further?" runner = self.__get_agent(session_id) try: response = runner.invoke({input: message}) except Exception as ex: self.__logger.exception(str(ex)) return "Sorry, please try again." return response if response else "No response received" def __get_history(self, session_id) -> ConversationBufferWindowMemory: if session_id not in self.__history: self.__history[session_id] = ConversationBufferWindowMemory(k=10, memory_key=chat_history, return_messages=True) return self.__history[session_id] session_id = str(uuid4()) response = send_message(session_id, "Hi") ``` ### Error Message and Stack Trace (if applicable) 2024-04-11 08:40:28,242] ERROR RhinoAgent 2 validation errors for AIMessage content str type expected (type=type_error.str) content value is not a valid list (type=type_error.list) Traceback (most recent call last): File "/Users/vmac/vWork/protoype/venv/lib/python3.11/site-packages/langchain/chains/base.py", line 163, in invoke raise e File "/Users/vmac/vWork/protoype/venv/lib/python3.11/site-packages/langchain/chains/base.py", line 158, in invoke final_outputs: Dict[str, Any] = self.prep_outputs( ^^^^^^^^^^^^^^^^^^ File "/Users/vmac/vWork/protoype/venv/lib/python3.11/site-packages/langchain/chains/base.py", line 455, in prep_outputs self.memory.save_context(inputs, outputs) File "/Users/vmac/vWork/protoype/venv/lib/python3.11/site-packages/langchain/memory/chat_memory.py", line 40, in save_context [HumanMessage(content=input_str), AIMessage(content=output_str)] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/vmac/vWork/protoype/venv/lib/python3.11/site-packages/langchain_core/messages/base.py", line 45, in __init__ return super().__init__(content=content, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/vmac/vWork/protoype/venv/lib/python3.11/site-packages/langchain_core/load/serializable.py", line 120, in __init__ super().__init__(**kwargs) File "/Users/vmac/vWork/protoype/venv/lib/python3.11/site-packages/pydantic/v1/main.py", line 341, in __init__ raise validation_error pydantic.v1.error_wrappers.ValidationError: 2 validation errors for AIMessage content str type expected (type=type_error.str) content value is not a valid list (type=type_error.list) ### Description I am trying to use langchain to build a basic chatbot. I have the prompt template defined and I am using openAI GPT4. When I ask any simple question that gets a string response from agent, it gives me the error above. Sometimes it also goes into a loop until I run into the "RateLimit" Error . Attached a screenshot for the same  ### System Info langchain==0.1.13 langchain-community==0.0.29 langchain-core==0.1.33 langchain-openai==0.1.1 langchain-text-splitters==0.0.1

LangChain throwing parsing error and goes in loop when returning a String response

https://api.github.com/repos/langchain-ai/langchain/issues/20341/comments

2024-04-11T17:24:10Z

2024-07-18T16:09:39Z

https://github.com/langchain-ai/langchain/issues/20341

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code The following code raises a malformed input request based on extraneous key [tools] when querying the AWS Bedrock service. ```python def init_test_db(debug = False): # Initialize the database # Load environment variables from .env.local file load_dotenv(dotenv_path='.env.local') # Read PostgreSQL connection parameters from environment variables POSTGRES_DB = os.getenv("POSTGRES_DB") POSTGRES_USER = os.getenv("POSTGRES_USER") POSTGRES_PASSWORD = os.getenv("POSTGRES_PASSWORD") POSTGRES_HOST = os.getenv("POSTGRES_HOST") POSTGRES_PORT = os.getenv("POSTGRES_PORT") # Build the PostgreSQL connection URI postgres_uri = f"postgresql://{POSTGRES_USER}:{POSTGRES_PASSWORD}@{POSTGRES_HOST}:{POSTGRES_PORT}/{POSTGRES_DB}" db = SQLDatabase.from_uri(postgres_uri) if debug: print(db.table_info) return db def init_llm(model = 'claude-v2.1'): # Create the LLM models = { 'claude-v2.1': 'anthropic.claude-v2:1', 'claude-v3-sonnet': 'anthropic.claude-3-sonnet-20240229-v1:0', 'mistral-large': "mistral.mistral-large-2402-v1:0", 'claude-v2': 'anthropic.claude-v2', 'jurassic-2-mid': 'ai21.j2-mid-v1', 'jurassic-2-ultra': "ai21.j2-ultra-v1", } llm = Bedrock(model_id=models[model], streaming=False, region_name='us-east-1') return llm llm = init_llm() db=init_test_db() agent_executer = create_sql_agent(llm=llm, db=db, agent_type="openai-tools", verbose=True) agent_executer.invoke("Which customer has the highest outstanding loan value") ``` ### Error Message and Stack Trace (if applicable) ValueError: Error raised by bedrock service: An error occurred (ValidationException) when calling the InvokeModelWithResponseStream operation: Malformed input request: #: extraneous key [tools] is not permitted, please reformat your input and try again. ### Description I am trying to use the langchain library to build an AWS bedrock powered SQL agent. ### System Info langchain==0.1.13 langchain-community==0.0.29 langchain-core==0.1.36 langchain-openai==0.1.1 langchain-text-splitters==0.0.1 Windows 11 Python 3.8.10

AWS Bedrock Service Raising Error: "Malformed input request: #: extraneous key [tools] is not permitted."

https://api.github.com/repos/langchain-ai/langchain/issues/20320/comments

2024-04-11T13:24:11Z

2024-07-03T07:41:40Z

https://github.com/langchain-ai/langchain/issues/20320

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code  ### Error Message and Stack Trace (if applicable) _No response_ ### Description i am trying to use gemini model from langchain CHatGoogleGenerativeAi, but i am getting this error ``` 'NoneType' object does not support item assignment (type=type_error) ``` ### System Info kaggle

Not able to use gemini from langchain in kaggle

https://api.github.com/repos/langchain-ai/langchain/issues/20317/comments

2024-04-11T11:18:54Z

2024-07-21T16:06:55Z

https://github.com/langchain-ai/langchain/issues/20317

_🤖_ <!-- Greeting --> Hey @nithinreddyyyyyy, back for more adventures in code, I see! Always a pleasure to dive into these challenges with you. 🚀 <!-- Answer --> Based on the information you provided and the similar issue I found in the LangChain repository, you can use FAISS for ParentDocumentRetriever in the LangChain framework as follows: ```python from langchain_community.vectorstores.faiss import FAISS from langchain_community.embeddings.openai import OpenAIEmbeddings from langchain_community.textsplitters import RecursiveCharacterTextSplitter from langchain.retrievers import ParentDocumentRetriever # Initialize the embeddings and FAISS vector store embeddings = OpenAIEmbeddings() faiss = FAISS.from_texts(texts, embeddings) # Define the child and parent splitters child_splitter = RecursiveCharacterTextSplitter(chunk_size=400, chunk_overlap=50) parent_splitter = RecursiveCharacterTextSplitter(chunk_size=1200, chunk_overlap=300) # Initialize the ParentDocumentRetriever with FAISS parent_document_retriever = ParentDocumentRetriever( vectorstore=faiss, docstore=store, child_splitter=child_splitter, parent_splitter=parent_splitter ) # Add documents to the retriever parent_document_retriever.add_documents(document, ids=None) ``` This code snippet includes the use of `child_splitter` and `parent_splitter` with FAISS as requested by the user. You can find more details in the [related issue](https://github.com/langchain-ai/langchain/issues/17237). For retrieving relevant documents with additional metadata and filtering capabilities, you can follow the pattern shown in the test cases for the FAISS class: ```python from langchain_community.vectorstores.faiss import FAISS from langchain_core.documents import Document # Assuming you have a list of texts and corresponding embeddings texts = ["text1", "text2", "text3"] metadatas = [{"page": i} for i in range(len(texts))] embeddings = [embedding1, embedding2, embedding3] # replace with actual embeddings # Create a FAISS instance from the texts, embeddings, and metadatas docsearch = FAISS.from_texts(texts, embeddings, metadatas=metadatas) # Now you can use this FAISS instance to retrieve relevant documents with filtering query = "some query" filter_criteria = {"page": 1} # or use a lambda function for more complex filtering relevant_docs = docsearch.similarity_search(query, k=1, filter=filter_criteria) ``` In this example, `relevant_docs` will contain the most relevant document to the query that also matches the filter criteria. You can adjust the `k` parameter to retrieve more documents and the `filter` parameter to apply different filtering rules. For asynchronous use cases, you can use the `await FAISS.afrom_texts` and `await docsearch.asimilarity_search` methods with the same parameters. You can find more details in the [test cases for the FAISS class](https://github.com/langchain-ai/langchain/blob/master/libs/community/tests/unit_tests/vectorstores/test_faiss.py). The `FAISS.from_texts` method in the LangChain framework is a class method that constructs a FAISS (Facebook AI Similarity Search) wrapper from raw documents. This method is a user-friendly interface that embeds documents, creates an in-memory docstore, and initializes the FAISS database. It is intended to be a quick way to get started with the framework. You can find more details in the [FAISS class source code](https://github.com/langchain-ai/langchain/blob/master/libs/community/langchain_community/vectorstores/faiss.py). I hope this helps! If you have any further questions, feel free to ask. ## Sources - [can i use FAISS isntead of Chroma for ParentDocumentRetriver?](https://github.com/langchain-ai/langchain/issues/17237) - [libs/community/tests/unit_tests/vectorstores/test_faiss.py](https://github.com/langchain-ai/langchain/blob/master/libs/community/tests/unit_tests/vectorstores/test_faiss.py) - [libs/community/langchain_community/vectorstores/faiss.py](https://github.com/langchain-ai/langchain/blob/master/libs/community/langchain_community/vectorstores/faiss.py) <!-- Dosu Comment Footer --> <details open> <summary> #### About Dosu </summary> This response is meant to be useful and save you time. It is **not meant to be a precise solution**, but rather a starting point for your own research. Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant. If you want to continue the conversation, start your reply with @dosu-bot. </details> _Originally posted by @dosubot[bot] in https://github.com/langchain-ai/langchain/issues/17576#issuecomment-1946203996_

save ParentDocumentRetriever in local path _🤖_

https://api.github.com/repos/langchain-ai/langchain/issues/20315/comments

2024-04-11T09:57:45Z

2024-07-18T16:09:29Z

https://github.com/langchain-ai/langchain/issues/20315

### Checklist - [X] I added a very descriptive title to this issue. - [X] I included a link to the documentation page I am referring to (if applicable). ### Issue with current documentation: In the current documentation : [Constitutional Chain Documentation](https://python.langchain.com/docs/guides/productionization/safety/constitutional_chain/#custom-principles) Usage of Constitutional Principles inside a Langchain Agent is not provided. It only has information on LLMChain. I want to use it inside a conversational agent, because I also have to pass the custom tools and history of conversation into the agent. Currently I am using create_openai_tools_agent() to create the agent. ### Idea or request for content: _No response_

Using Constitutional AI with Lanchain Agents

https://api.github.com/repos/langchain-ai/langchain/issues/20312/comments

2024-04-11T09:22:25Z

2024-07-18T16:09:24Z

https://github.com/langchain-ai/langchain/issues/20312

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code INITIALIZE_AGENT ### Error Message and Stack Trace (if applicable) _No response_ ### Description Everytime, i use this node, i see it is up for deprecation in langchain's V2. Was wondering if it'll be too much to ask for this to not be deprecated but maintained, because it's the only agent executing function and creator agent that works with almost any LLM i multiswitch easily with e.g (Gemini, Openrouter, Huggingface) . The new create_openai_structured or react_agents are often problematic requiring a ton of time and wasted effort to debug. Even with langchain's examples, they often don't work. Just thought i would mention this. Thanks ### System Info INITIALIZE_AGENT

INITIALIZE_AGENT (LANGCHAIN'S TRADITIONAL AGENT EXECUTOR CREATOR)

https://api.github.com/repos/langchain-ai/langchain/issues/20309/comments

2024-04-11T06:44:41Z

2024-07-18T16:09:19Z

https://github.com/langchain-ai/langchain/issues/20309

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code Code that doesn't work: ``` retriever = self.vector_store.as_retriever(search_kwargs={ 'k': 1, 'filter': {'portfolio_uri':{'$eq': params.portfolio_uri}} }) docs = retriever.get_relevant_documents(query=query) ``` Code that works: ``` retriever = self.vector_store.as_retriever(search_kwargs={ 'k': limit, 'filter': {'portfolio_uri':{'$in': [params.portfolio_uri]}} }) docs = retriever.get_relevant_documents(query=query) ``` ### Error Message and Stack Trace (if applicable) File "/Users/anthonydemattos/syyclops/open-operator/.venv/lib/python3.10/site-packages/sqlalchemy/engine/base.py", line 1971, in _exec_single_context self.dialect.do_execute( File "/Users/anthonydemattos/syyclops/open-operator/.venv/lib/python3.10/site-packages/sqlalchemy/engine/default.py", line 919, in do_execute cursor.execute(statement, parameters) sqlalchemy.exc.ProgrammingError: (psycopg2.errors.UndefinedFunction) function jsonb_path_match(json, unknown, unknown) does not exist LINE 3: ...2edf-c5b6-4a08-9a06-c841d8c6dab9'::uuid::UUID AND jsonb_path... ^ HINT: No function matches the given name and argument types. You might need to add explicit type casts. [SQL: SELECT langchain_pg_embedding.collection_id AS langchain_pg_embedding_collection_id, langchain_pg_embedding.embedding AS langchain_pg_embedding_embedding, langchain_pg_embedding.document AS langchain_pg_embedding_document, langchain_pg_embedding.cmetadata AS langchain_pg_embedding_cmetadata, langchain_pg_embedding.custom_id AS langchain_pg_embedding_custom_id, langchain_pg_embedding.uuid AS langchain_pg_embedding_uuid, langchain_pg_embedding.embedding <=> %(embedding_1)s AS distance FROM langchain_pg_embedding JOIN langchain_pg_collection ON langchain_pg_embedding.collection_id = langchain_pg_collection.uuid WHERE langchain_pg_embedding.collection_id = %(collection_id_1)s::UUID AND jsonb_path_match(langchain_pg_embedding.cmetadata, %(jsonb_path_match_1)s, %(jsonb_path_match_2)s) ORDER BY distance ASC LIMIT %(param_1)s] [parameters: {'embedding_1': '[-0.010107061089572959,-0.013629535476347075,-0.0012468165027500532,-0.024999106785678033,-0.03649423341900812,0.012025240109652443,-0.03317403857405 ... (32593 characters truncated) ... .024217885455325203,-0.029881744291334902,0.015178028787260737,-0.01019773838247912,-0.006612486799366736,-0.02557107296798936,-0.019600304222168217]', 'collection_id_1': UUID('622b2edf-c5b6-4a08-9a06-c841d8c6dab9'), 'jsonb_path_match_1': '$.portfolio_uri == $value', 'jsonb_path_match_2': '{"value": "https://syyclops.com/example"}', 'param_1': 1}] (Background on this error at: https://sqlalche.me/e/20/f405) ### Description I am trying to do a vector store similarity search with pg vector and provide a filter of the metadata. For some reason when i do $eq it doesn't work but if I do $in: [.. then it does work ### System Info python -m langchain_core.sys_info System Information ------------------ > OS: Darwin > OS Version: Darwin Kernel Version 23.0.0: Fri Sep 15 14:41:43 PDT 2023; root:xnu-10002.1.13~1/RELEASE_ARM64_T6000 > Python Version: 3.10.6 | packaged by conda-forge | (main, Aug 22 2022, 20:38:29) [Clang 13.0.1 ] Package Information ------------------- > langchain_core: 0.1.41 > langchain: 0.1.12 > langchain_community: 0.0.32 > langsmith: 0.1.43 > langchain_openai: 0.1.2 > langchain_text_splitters: 0.0.1 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve

PGVector filter $eq

https://api.github.com/repos/langchain-ai/langchain/issues/20293/comments

2024-04-10T19:32:41Z

2024-04-18T20:43:28Z

https://github.com/langchain-ai/langchain/issues/20293

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code `%pip install --upgrade --quiet azure-search-documents %pip install --upgrade --quiet azure-identity import os from langchain_community.vectorstores.azuresearch import AzureSearch from langchain_openai import AzureOpenAIEmbeddings, OpenAIEmbeddings # Option 2: use an Azure OpenAI account with a deployment of an embedding model azure_endpoint: str = "PLACEHOLDER FOR YOUR AZURE OPENAI ENDPOINT" azure_openai_api_key: str = "PLACEHOLDER FOR YOUR AZURE OPENAI KEY" azure_openai_api_version: str = "2023-05-15" azure_deployment: str = "text-embedding-ada-002" vector_store_address: str = "YOUR_AZURE_SEARCH_ENDPOINT" vector_store_password: str = "YOUR_AZURE_SEARCH_ADMIN_KEY" # Option 2: Use AzureOpenAIEmbeddings with an Azure account embeddings: AzureOpenAIEmbeddings = AzureOpenAIEmbeddings( azure_deployment=azure_deployment, openai_api_version=azure_openai_api_version, azure_endpoint=azure_endpoint, api_key=azure_openai_api_key, ) index_name: str = "langchain-vector-demo" vector_store: AzureSearch = AzureSearch( azure_search_endpoint=vector_store_address, azure_search_key=vector_store_password, index_name=index_name, embedding_function=embeddings.embed_query, ) from langchain.text_splitter import ( CharacterTextSplitter, RecursiveCharacterTextSplitter, ) from langchain.document_loaders import DirectoryLoader, PyPDFLoader # Read the PDF file using the langchain loader pdf_link = "test.pdf" loader = PyPDFLoader(pdf_link, extract_images=False) data = loader.load_and_split() text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0) docs = text_splitter.split_documents(data) vector_store.add_documents(documents=docs)` ### Error Message and Stack Trace (if applicable) --------------------------------------------------------------------------- AttributeError Traceback (most recent call last) Cell In[12], line 2 1 for i in range(0, len(docs)): ----> 2 vector_store.add_documents(documents=docs[i]) 3 time.sleep(5) File ~/anaconda3/envs/rag_azure/lib/python3.10/site-packages/langchain_core/vectorstores.py:136, in VectorStore.add_documents(self, documents, **kwargs) 127 """Run more documents through the embeddings and add to the vectorstore. 128 129 Args: (...) 133 List[str]: List of IDs of the added texts. 134 """ 135 # TODO: Handle the case where the user doesn't provide ids on the Collection --> 136 texts = [doc.page_content for doc in documents] 137 metadatas = [doc.metadata for doc in documents] 138 return self.add_texts(texts, metadatas, **kwargs) File ~/anaconda3/envs/rag_azure/lib/python3.10/site-packages/langchain_core/vectorstores.py:136, in <listcomp>(.0) 127 """Run more documents through the embeddings and add to the vectorstore. 128 129 Args: (...) 133 List[str]: List of IDs of the added texts. 134 """ 135 # TODO: Handle the case where the user doesn't provide ids on the Collection --> 136 texts = [doc.page_content for doc in documents] 137 metadatas = [doc.metadata for doc in documents] 138 return self.add_texts(texts, metadatas, **kwargs) AttributeError: 'tuple' object has no attribute 'page_content' ### Description I am using langchain to connect to Azure AI Search and create vector stores and add documents to them so I can create a RAG application. I tried to replicate the notebook provided by Langchain for Azure AI Search [https://python.langchain.com/docs/integrations/vectorstores/azuresearch/](url) but its failing with the above error I do see page_content in 'docs' so I am not sure where is the problem. I got langchain_core.documents.base.Document on type(docs[0]) Here is an example of how one of the element of the doc looks print(docs[5]) Document(page_content='Modify the likelihood of specified tokens appearing in the completion. Accepts a json object that maps tokens (specified by their token ID in the GPT tokenizer) to an associated bias value from -100 to 100. You can use this tokenizer tool (which works for both GPT-2 and GPT-3) to convert text to token IDs. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect varies per model, but values between -1 and 1 should decrease or increase likelihood of selection', metadata={'source': 'test.pdf', 'page': 3}) ### System Info platform - mac python - 3.10 langchain==0.1.15 langchain-community==0.0.32 langchain-core==0.1.41 langchain-openai==0.0.2.post1 langchain-text-splitters==0.0.1

Error when adding documents to vector_store - Azure AI Search

https://api.github.com/repos/langchain-ai/langchain/issues/20283/comments

2024-04-10T15:45:26Z

2024-07-19T16:08:10Z

https://github.com/langchain-ai/langchain/issues/20283

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python import time from langchain.graphs import Neo4jGraph from langchain_openai import AzureChatOpenAI from langchain.prompts.prompt import PromptTemplate from langchain.chains import GraphCypherQAChain llm=AzureChatOpenAI(azure_deployment=MODEL_CHAT, model_name=MODEL_CHAT, azure_endpoint=API_ENDPOINT, openai_api_version=API_VERSION, openai_api_key=API_KEY, temperature=0, streaming=True) neo4j_graph = Neo4jGraph(url=NEO4J_URI, username=NEO4J_USERNAME, password=NEO4J_PASSWORD) CYPHER_GENERATION_TEMPLATE = """You are an expert Neo4j Cypher translator who understands the question in english and convert to Cypher strictly based on the Neo4j Schema provided and following the instructions below: <instructions> * Use aliases to refer the node or relationship in the generated Cypher query * Generate Cypher query compatible ONLY for Neo4j Version 5 * Do not use EXISTS, SIZE keywords in the cypher. Use alias when using the WITH keyword * Use only Nodes and relationships mentioned in the schema * Always enclose the Cypher output inside 3 backticks (```) * Always do a case-insensitive and fuzzy search for any properties related search. Eg: to search for a Person name use `toLower(p.name) contains 'neo4j'` * Cypher is NOT SQL. So, do not mix and match the syntaxes </instructions> Strictly use this Schema for Cypher generation: <schema> {schema} </schema> The samples below follow the instructions and the schema mentioned above. So, please follow the same when you generate the cypher: <samples> Human: Which manager manages most people directly? How many employees? Assistant: ```MATCH (p:Person)-[r:IS_MANAGER_OF]->() WITH p, COUNT(r) AS NumberOfEmployees ORDER BY NumberOfEmployees DESC RETURN p.name, NumberOfEmployees LIMIT 1``` </samples> Human: {question} Assistant: """ CYPHER_GENERATION_PROMPT = PromptTemplate(input_variables=['schema','question'], validate_template=True, template=CYPHER_GENERATION_TEMPLATE) chain = GraphCypherQAChain.from_llm( llm, graph=neo4j_graph, cypher_prompt=CYPHER_GENERATION_PROMPT, validate_cypher=True, return_intermediate_steps=True ) question="Who should we fire from the Example department?" cypher_cmd=chain.invoke(question) ``` ### Error Message and Stack Trace (if applicable) ```bash --------------------------------------------------------------------------- CypherSyntaxError Traceback (most recent call last) File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/langchain_community/graphs/neo4j_graph.py:164](.pyenv/versions/3.11.7/lib/python3.11/site-packages/langchain_community/graphs/neo4j_graph.py#line=163), in Neo4jGraph.query(self, query, params) 163 try: --> 164 data = session.run(Query(text=query, timeout=self.timeout), params) 165 json_data = [r.data() for r in data] File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/work/session.py:313](.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/work/session.py#line=312), in Session.run(self, query, parameters, **kwargs) 312 parameters = dict(parameters or {}, **kwargs) --> 313 self._auto_result._run( 314 query, parameters, self._config.database, 315 self._config.impersonated_user, self._config.default_access_mode, 316 bookmarks, self._config.notifications_min_severity, 317 self._config.notifications_disabled_categories, 318 ) 320 return self._auto_result File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/work/result.py:181](.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/work/result.py#line=180), in Result._run(self, query, parameters, db, imp_user, access_mode, bookmarks, notifications_min_severity, notifications_disabled_categories) 180 self._connection.send_all() --> 181 self._attach() File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/work/result.py:301](.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/work/result.py#line=300), in Result._attach(self) 300 while self._attached is False: --> 301 self._connection.fetch_message() File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/io/_common.py:178](.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/io/_common.py#line=177), in ConnectionErrorHandler.__getattr__.<locals>.outer.<locals>.inner(*args, **kwargs) 177 try: --> 178 func(*args, **kwargs) 179 except (Neo4jError, ServiceUnavailable, SessionExpired) as exc: File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/io/_bolt.py:849](.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/io/_bolt.py#line=848), in Bolt.fetch_message(self) 846 tag, fields = self.inbox.pop( 847 hydration_hooks=self.responses[0].hydration_hooks 848 ) --> 849 res = self._process_message(tag, fields) 850 self.idle_since = monotonic() File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/io/_bolt5.py:369](.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/io/_bolt5.py#line=368), in Bolt5x0._process_message(self, tag, fields) 368 try: --> 369 response.on_failure(summary_metadata or {}) 370 except (ServiceUnavailable, DatabaseUnavailable): File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/io/_common.py:245](.pyenv/versions/3.11.7/lib/python3.11/site-packages/neo4j/_sync/io/_common.py#line=244), in Response.on_failure(self, metadata) 244 Util.callback(handler) --> 245 raise Neo4jError.hydrate(**metadata) CypherSyntaxError: {code: Neo.ClientError.Statement.SyntaxError} {message: Invalid input 'I': expected "ALTER" "CALL" "CREATE" "DEALLOCATE" "DELETE" "DENY" "DETACH" "DROP" "DRYRUN" "ENABLE" "FOREACH" "GRANT" "LOAD" "MATCH" "MERGE" "NODETACH" "OPTIONAL" "REALLOCATE" "REMOVE" "RENAME" "RETURN" "REVOKE" "SET" "SHOW" "START" "STOP" "TERMINATE" "UNWIND" "USE" "USING" "WITH" (line 1, column 1 (offset: 0)) "I'm sorry, I cannot generate a query for this question as it goes against ethical and moral principles. It is not appropriate to use data and technology to harm or discriminate against individuals." ^} During handling of the above exception, another exception occurred: ValueError Traceback (most recent call last) Cell In[7], line 2 1 question="Who should we fire from 91130 Veh Verif & Value Confirmation?" ----> 2 cypher_cmd=chain.invoke(question) File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/langchain/chains/base.py:162](.pyenv/versions/3.11.7/lib/python3.11/site-packages/langchain/chains/base.py#line=161), in Chain.invoke(self, input, config, **kwargs) 160 except BaseException as e: 161 run_manager.on_chain_error(e) --> 162 raise e 163 run_manager.on_chain_end(outputs) 164 final_outputs: Dict[str, Any] = self.prep_outputs( 165 inputs, outputs, return_only_outputs 166 ) File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/langchain/chains/base.py:156](.pyenv/versions/3.11.7/lib/python3.11/site-packages/langchain/chains/base.py#line=155), in Chain.invoke(self, input, config, **kwargs) 149 run_manager = callback_manager.on_chain_start( 150 dumpd(self), 151 inputs, 152 name=run_name, 153 ) 154 try: 155 outputs = ( --> 156 self._call(inputs, run_manager=run_manager) 157 if new_arg_supported 158 else self._call(inputs) 159 ) 160 except BaseException as e: 161 run_manager.on_chain_error(e) File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/langchain/chains/graph_qa/cypher.py:267](.pyenv/versions/3.11.7/lib/python3.11/site-packages/langchain/chains/graph_qa/cypher.py#line=266), in GraphCypherQAChain._call(self, inputs, run_manager) 264 # Retrieve and limit the number of results 265 # Generated Cypher be null if query corrector identifies invalid schema 266 if generated_cypher: --> 267 context = self.graph.query(generated_cypher)[: self.top_k] 268 else: 269 context = [] File [~/.pyenv/versions/3.11.7/lib/python3.11/site-packages/langchain_community/graphs/neo4j_graph.py:170](.pyenv/versions/3.11.7/lib/python3.11/site-packages/langchain_community/graphs/neo4j_graph.py#line=169), in Neo4jGraph.query(self, query, params) 168 return json_data 169 except CypherSyntaxError as e: --> 170 raise ValueError(f"Generated Cypher Statement is not valid\n{e}") ValueError: Generated Cypher Statement is not valid {code: Neo.ClientError.Statement.SyntaxError} {message: Invalid input 'I': expected "ALTER" "CALL" "CREATE" "DEALLOCATE" "DELETE" "DENY" "DETACH" "DROP" "DRYRUN" "ENABLE" "FOREACH" "GRANT" "LOAD" "MATCH" "MERGE" "NODETACH" "OPTIONAL" "REALLOCATE" "REMOVE" "RENAME" "RETURN" "REVOKE" "SET" "SHOW" "START" "STOP" "TERMINATE" "UNWIND" "USE" "USING" "WITH" (line 1, column 1 (offset: 0)) "I'm sorry, I cannot generate a query for this question as it goes against ethical and moral principles. It is not appropriate to use data and technology to harm or discriminate against individuals." ^} ``` ### Description GraphCypherQAChain is trying to generate a Cypher query from the LLM's error message: "I'm sorry, I cannot generate a query for this question as it goes against ethical and moral principles. It is not appropriate to use data and technology to harm or discriminate against individuals." This code works for other prompts except for those containing sensitive questions or information outside the provided schema. ### System Info System Information ------------------ > OS: Linux > OS Version: #104-Ubuntu SMP Tue Jan 9 15:25:40 UTC 2024 > Python Version: 3.11.7 (main, Feb 15 2024, 09:21:29) [Clang 14.0.0 ] Package Information ------------------- > langchain_core: 0.1.30 > langchain: 0.1.7 > langchain_community: 0.0.20 > langsmith: 0.1.22 > langchain_cli: 0.0.21 > langchain_openai: 0.0.8 > langserve: 0.0.41 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph

GraphCypherQAChain tries to create a query from a sensitive question.

https://api.github.com/repos/langchain-ai/langchain/issues/20280/comments

2024-04-10T15:03:20Z

2024-07-17T16:06:18Z

https://github.com/langchain-ai/langchain/issues/20280

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code 1 model = AzureChatOpenAI( openai_api_version="2023-03-15-preview", azure_deployment="gpt_chat_updated", # in Azure, this deployment has version 0613 - input and output tokens are counted separately ) message = HumanMessage( content='Where is whitehouse located?' ) with get_openai_callback() as cb: response = model([message]) print(f"AzureChatAPI response {response}") print( f"Total Cost (USD): ${format(cb.total_cost, '.6f')}" ) #### output AzureChatAPI response content='The White House is located in Washington, D.C., the capital city of the United States.' response_metadata={'token_usage': {'completion_tokens': 19, 'prompt_tokens': 13, 'total_tokens': 32}, 'model_name': 'gpt-35-turbo-16k', 'system_fingerprint': None, 'finish_reason': 'stop', 'logprobs': None, 'content_filter_results': {}} id='run-7976ac34-171f-498e-8819-3b67fbd3f21e-0' **Total Cost (USD): $0.000115** # ### Example Code 2 chatlitellm_model = ChatLiteLLM(model="azure/gpt_chat_updated") with get_openai_callback() as cb: response = chatlitellm_model([message]) print(f"Chatlite Response {response}") print( f"Total Cost (USD): ${format(cb.total_cost, '.6f')}" ) #### output Chatlite Response content='The White House is located at 1600 Pennsylvania Avenue NW, Washington, D.C., United States.' response_metadata={'token_usage': Usage(completion_tokens=21, prompt_tokens=13, total_tokens=34), 'model': 'azure/gpt_chat_updated', 'finish_reason': 'stop'} id='run-f98efc18-6c9f-4b61-acfa-4baf818ab38d-0' **Total Cost (USD): $0.000000** ### Error Message and Stack Trace (if applicable) Please note the difference in model_name and hence the Total Cost!!! ### Description get_openai_callback() returns the incorrect model_name (as is deployment name on azure), hence blocking the Cost instrumentation wihile using ChatliteLLM though the same model version works perfectly fine when used with AzureChatOpenAI. This is further leading to reuturning 0 cost by get_openai_callback() though the original model name and version is listed in openai_info.py. ### System Info langchain==0.1.12 langchain-community==0.0.28 langchain-core==0.1.41 langchain-openai==0.1.2 langchain-text-splitters==0.0.1

Altered model name with ChatLiteLLM hindering cost instrumentation with get_openai_callback

https://api.github.com/repos/langchain-ai/langchain/issues/20277/comments

2024-04-10T14:39:57Z

2024-07-17T16:06:14Z

https://github.com/langchain-ai/langchain/issues/20277

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code The following code: ''' tools = load_tools(["llm-math"],llm=llm) chain = ConversationalRetrievalChain.from_llm( llm=llm, tools=tools, ''' ### Error Message and Stack Trace (if applicable) causes this error: \venv\lib\site-packages\langchain_core\load\serializable.py", line 120, in __init__ super().__init__(**kwargs) File "pydantic\main.py", line 341, in pydantic.main.BaseModel.__init__ pydantic.error_wrappers.ValidationError: 1 validation error for ConversationalRetrievalChain tool extra fields not permitted (type=value_error.extra) ### Description I'm trying to use tools with ConversationalRetrievalChain.from_llm and this error occurs pydantic.error_wrappers.ValidationError: 1 validation error for ConversationalRetrievalChain tool ### System Info aiohttp==3.8.4 aiosignal==1.3.1 altair==5.0.1 anyio==3.7.0 async-timeout==4.0.2 attrs==23.1.0 backoff==2.2.1 beautifulsoup4==4.12.2 blinker==1.6.2 bs4==0.0.1 cachetools==5.3.1 certifi==2023.5.7 cffi==1.15.1 chardet==5.1.0 charset-normalizer==3.1.0 chromadb==0.3.26 click==8.1.3 clickhouse-connect==0.6.4 colorama==0.4.6 coloredlogs==15.0.1 cryptography==41.0.2 dataclasses-json==0.5.9 decorator==5.1.1 distro==1.9.0 dnspython==2.3.0 docopt==0.6.2 document-utils==1.8.1 duckdb==0.8.1 et-xmlfile==1.1.0 exceptiongroup==1.1.2 faiss-cpu==1.7.4 fastapi==0.99.1 filetype==1.2.0 flatbuffers==23.5.26 frozenlist==1.3.3 gitdb==4.0.10 GitPython==3.1.31 greenlet==2.0.2 h11==0.14.0 hnswlib==0.7.0 httpcore==1.0.3 httptools==0.5.0 httpx==0.26.0 humanfriendly==10.0 idna==3.4 importlib-metadata==6.7.0 Jinja2==3.1.2 joblib==1.3.1 jsonpatch==1.33 jsonpointer==2.4 jsonschema==4.17.3 langchain==0.1.15 langchain-community==0.0.32 langchain-core==0.1.41 langchain-experimental==0.0.47 langchain-text-splitters==0.0.1 langchainplus-sdk==0.0.20 langsmith==0.1.43 libmagic==1.0 loguru==0.7.0 lxml==4.9.3 lz4==4.3.2 Markdown==3.4.4 markdown-it-py==3.0.0 MarkupSafe==2.1.3 marshmallow==3.19.0 marshmallow-enum==1.5.1 mdurl==0.1.2 monotonic==1.6 mpmath==1.3.0 msg-parser==1.2.0 multidict==6.0.4 mypy-extensions==1.0.0 nltk==3.8.1 numexpr==2.8.4 numpy==1.25.0 olefile==0.46 onnxruntime==1.15.1 openai==0.27.8 openapi-schema-pydantic==1.2.4 openpyxl==3.1.2 orjson==3.10.0 overrides==7.3.1 packaging==23.2 pandas==2.0.3 pdf2image==1.16.3 pdfminer.six==20221105 Pillow==9.5.0 pinecone-client==2.2.2 pipreqs==0.4.13 posthog==3.0.1 protobuf==4.23.3 pulsar-client==3.2.0 pyarrow==12.0.1 pycparser==2.21 pycryptodome==3.18.0 pydantic==1.10.10 pydeck==0.8.1b0 Pygments==2.15.1 Pympler==1.0.1 pypandoc==1.11 pypdf==3.12.0 pyreadline3==3.4.1 pyrsistent==0.19.3 python-dateutil==2.8.2 python-docx==0.8.11 python-dotenv==1.0.0 python-magic==0.4.27 python-magic-bin==0.4.14 python-pptx==0.6.21 pytz==2023.3 pytz-deprecation-shim==0.1.0.post0 PyYAML==6.0 regex==2023.6.3 requests==2.31.0 rich==13.4.2 six==1.16.0 smmap==5.0.0 sniffio==1.3.0 soupsieve==2.5 SQLAlchemy==2.0.17 starlette==0.27.0 streamlit==1.24.0 sympy==1.12 tabulate==0.9.0 tenacity==8.2.2 tiktoken==0.4.0 tokenizers==0.13.3 toml==0.10.2 toolz==0.12.0 tornado==6.3.2 tqdm==4.65.0 typing-inspect==0.9.0 typing_extensions==4.9.0 tzdata==2023.3 tzlocal==4.3.1 unstructured==0.8.1 urllib3==2.0.3 uvicorn==0.22.0 validators==0.20.0 vectorhub==1.8.3 watchdog==3.0.0 watchfiles==0.19.0 websockets==11.0.3 win32-setctime==1.1.0 xlrd==2.0.1 XlsxWriter==3.1.2 yarg==0.1.9 yarl==1.9.2 zipp==3.15.0 zstandard==0.21.0

Adding tools to ConversationalRetrievalChain.from_llm causes Pydantic error

https://api.github.com/repos/langchain-ai/langchain/issues/20276/comments

2024-04-10T14:04:46Z

2024-07-18T16:09:09Z

https://github.com/langchain-ai/langchain/issues/20276

### Checklist - [X] I added a very descriptive title to this issue. - [X] I included a link to the documentation page I am referring to (if applicable). ### Issue with current documentation: The documentation at https://python.langchain.com/docs/modules/model_io/output_parsers/types/json/ seems to imply that the generated output is validated against the Pydantic schema and only JSON objects that conform to the schema are returned. However, as I understand the source code, this does not seem to be the case. Please consider adding a note to the JSON parser documentation that one should use PydanticOutputParser, if validation against the schema is desired. ### Idea or request for content: _No response_

DOC: JsonOutputParser does not mention that input is not validated against schema.

https://api.github.com/repos/langchain-ai/langchain/issues/20266/comments

2024-04-10T10:22:42Z

2024-07-18T16:09:04Z

https://github.com/langchain-ai/langchain/issues/20266

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python from langchain import hub from langchain.agents import AgentExecutor, create_openai_tools_agent from langchain_community.tools.tavily_search import TavilySearchResults from langchain_openai import ChatOpenAI tools = [TavilySearchResults(max_results=1)] prompt = hub.pull("hwchase17/openai-tools-agent") llm = ChatOpenAI(model="gpt-4-turbo-2024-04-09", temperature=0) agent = create_openai_tools_agent(llm, tools, prompt) agent_executor = AgentExecutor(agent=agent, tools=tools) result = agent_executor.invoke({"input": "What's the weather like in Tokyo?"}) print(result["output"]) ``` ### Error Message and Stack Trace (if applicable) ``` Traceback (most recent call last): File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/src/langchain_agent.py", line 20, in <module> result = agent_executor.invoke({"input": "What's the weather like in Tokyo?"}) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain/chains/base.py", line 163, in invoke raise e File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain/chains/base.py", line 153, in invoke self._call(inputs, run_manager=run_manager) File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain/agents/agent.py", line 1432, in _call next_step_output = self._take_next_step( ^^^^^^^^^^^^^^^^^^^^^ File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain/agents/agent.py", line 1138, in _take_next_step [ File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain/agents/agent.py", line 1138, in <listcomp> [ File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain/agents/agent.py", line 1166, in _iter_next_step output = self.agent.plan( ^^^^^^^^^^^^^^^^ File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain/agents/agent.py", line 514, in plan for chunk in self.runnable.stream(inputs, config={"callbacks": callbacks}): File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain_core/runnables/base.py", line 2875, in stream yield from self.transform(iter([input]), config, **kwargs) File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain_core/runnables/base.py", line 2862, in transform yield from self._transform_stream_with_config( File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain_core/runnables/base.py", line 1880, in _transform_stream_with_config chunk: Output = context.run(next, iterator) # type: ignore ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain_core/runnables/base.py", line 2826, in _transform for output in final_pipeline: File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain_core/runnables/base.py", line 1283, in transform for chunk in input: File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain_core/runnables/base.py", line 4722, in transform yield from self.bound.transform( File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain_core/runnables/base.py", line 1300, in transform yield from self.stream(final, config, **kwargs) File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py", line 245, in stream raise e File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py", line 225, in stream for chunk in self._stream(messages, stop=stop, **kwargs): File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/langchain_openai/chat_models/base.py", line 460, in _stream with self.client.create(messages=message_dicts, **params) as response: ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/openai/_utils/_utils.py", line 275, in wrapper return func(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/openai/resources/chat/completions.py", line 667, in create return self._post( ^^^^^^^^^^^ File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/openai/_base_client.py", line 1213, in post return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/openai/_base_client.py", line 902, in request return self._request( ^^^^^^^^^^^^^^ File "/home/oshima/work/src/os1ma/debug_langchain_gpt-4-turbo-2024-04-09/.venv/lib/python3.11/site-packages/openai/_base_client.py", line 993, in _request raise self._make_status_error_from_response(err.response) from None openai.BadRequestError: Error code: 400 - {'error': {'message': "Unknown parameter: 'messages[2].tool_calls[0].index'.", 'type': 'invalid_request_error', 'param': 'messages[2].tool_calls[0].index', 'code': 'unknown_parameter'}} ``` ### Description OpenAI tools agent initialized by `create_openai_tools_agent` raise error if use with gpt-4-turbo-2024-04-09. This error is not caused with other OpenAI model like gpt-4-0125-preview, gpt-3.5-turbo-0125. If I only use openai package without langchain, this error is not caused. ### System Info ``` $ python -m langchain_core.sys_info System Information ------------------ > OS: Linux > OS Version: #111~20.04.1-Ubuntu SMP Mon Mar 11 15:44:43 UTC 2024 > Python Version: 3.11.9 (main, Apr 10 2024, 18:31:06) [GCC 9.4.0] Package Information ------------------- > langchain_core: 0.1.41 > langchain: 0.1.15 > langchain_community: 0.0.32 > langsmith: 0.1.43 > langchain_openai: 0.1.2 > langchain_text_splitters: 0.0.1 > langchainhub: 0.1.15 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve ```

OpenAI tools agent error if use with gpt-4-turbo-2024-04-09

https://api.github.com/repos/langchain-ai/langchain/issues/20264/comments

2024-04-10T09:48:54Z

2024-04-13T06:23:46Z

https://github.com/langchain-ai/langchain/issues/20264

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code I checked first connection and the following code works ``` from langchain.graphs import Neo4jGraph from neo4j import GraphDatabase url = "neo4j://localhost:7687" username ="neo4j" password = <password> def _create_and_return_greeting(tx, message): result = tx.run("CREATE (a:Greeting) " "SET a.message = $message " "RETURN a.message + ', from node ' + id(a)", message=message) return result.single()[0] message="hello, world" driver=GraphDatabase.driver(url, auth=(username, password)) with driver.session() as session: greeting = session.execute_write(_create_and_return_greeting, message) print(greeting) ``` Hovewer trying to connect ``` graph = Neo4jGraph(url=url, username=username, password=password) ``` gives error ### Error Message and Stack Trace (if applicable) ``` --------------------------------------------------------------------------- CypherSyntaxError Traceback (most recent call last) [/usr/local/lib/python3.10/dist-packages/langchain_community/graphs/neo4j_graph.py](https://vscode-remote+dev-002dcontainer-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.vscode-resource.vscode-cdn.net/usr/local/lib/python3.10/dist-packages/langchain_community/graphs/neo4j_graph.py) in query(self, query, params) 245 try: --> 246 data = session.run(Query(text=query, timeout=self.timeout), params) 247 json_data = [r.data() for r in data] [/usr/local/lib/python3.10/dist-packages/neo4j/_sync/work/session.py](https://vscode-remote+dev-002dcontainer-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.vscode-resource.vscode-cdn.net/usr/local/lib/python3.10/dist-packages/neo4j/_sync/work/session.py) in run(self, query, parameters, **kwargs) 312 parameters = dict(parameters or {}, **kwargs) --> 313 self._auto_result._run( 314 query, parameters, self._config.database, [/usr/local/lib/python3.10/dist-packages/neo4j/_sync/work/result.py](https://vscode-remote+dev-002dcontainer-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.vscode-resource.vscode-cdn.net/usr/local/lib/python3.10/dist-packages/neo4j/_sync/work/result.py) in _run(self, query, parameters, db, imp_user, access_mode, bookmarks, notifications_min_severity, notifications_disabled_categories) 180 self._connection.send_all() --> 181 self._attach() 182 [/usr/local/lib/python3.10/dist-packages/neo4j/_sync/work/result.py](https://vscode-remote+dev-002dcontainer-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.vscode-resource.vscode-cdn.net/usr/local/lib/python3.10/dist-packages/neo4j/_sync/work/result.py) in _attach(self) 300 while self._attached is False: --> 301 self._connection.fetch_message() 302 [/usr/local/lib/python3.10/dist-packages/neo4j/_sync/io/_common.py](https://vscode-remote+dev-002dcontainer-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.vscode-resource.vscode-cdn.net/usr/local/lib/python3.10/dist-packages/neo4j/_sync/io/_common.py) in inner(*args, **kwargs) 177 try: --> 178 func(*args, **kwargs) 179 except (Neo4jError, ServiceUnavailable, SessionExpired) as exc: [/usr/local/lib/python3.10/dist-packages/neo4j/_sync/io/_bolt.py](https://vscode-remote+dev-002dcontainer-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.vscode-resource.vscode-cdn.net/usr/local/lib/python3.10/dist-packages/neo4j/_sync/io/_bolt.py) in fetch_message(self) 849 ) --> 850 res = self._process_message(tag, fields) 851 self.idle_since = monotonic() [/usr/local/lib/python3.10/dist-packages/neo4j/_sync/io/_bolt4.py](https://vscode-remote+dev-002dcontainer-002b7b22686f737450617468223a222f6d656469612f6a6d2f686464446174612f70726f6a656374732f746879726f69645f6b6e6f776c656467655f6772617068222c226c6f63616c446f636b6572223a66616c73652c22636f6e66696746696c65223a7b22246d6964223a312c22667350617468223a222f6d656469612f6a6d2f686464446174612f70726f6a656374732f746879726f69645f6b6e6f776c656467655f67726170682f2e646576636f6e7461696e65722f646576636f6e7461696e65722e6a736f6e222c2265787465726e616c223a2266696c653a2f2f2f6d656469612f6a6d2f686464446174612f70726f6a656374732f746879726f69645f6b6e6f776c656467655f67726170682f2e646576636f6e7461696e65722f646576636f6e7461696e65722e6a736f6e222c2270617468223a222f6d656469612f6a6d2f686464446174612f70726f6a656374732f746879726f69645f6b6e6f776c656467655f67726170682f2e646576636f6e7461696e65722f646576636f6e7461696e65722e6a736f6e222c22736368656d65223a2266696c65227d7d.vscode-resource.vscode-cdn.net/usr/local/lib/python3.10/dist-packages/neo4j/_sync/io/_bolt4.py) in _process_message(self, tag, fields) 367 try: --> 368 response.on_failure(summary_metadata or {}) 369 except (ServiceUnavailable, DatabaseUnavailable): [/usr/local/lib/python3.10/dist-packages/neo4j/_sync/io/_common.py](https://vscode-remote+dev-002dcontainer-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.vscode-resource.vscode-cdn.net/usr/local/lib/python3.10/dist-packages/neo4j/_sync/io/_common.py) in on_failure(self, metadata) 244 Util.callback(handler) --> 245 raise Neo4jError.hydrate(**metadata) 246 CypherSyntaxError: {code: Neo.ClientError.Statement.SyntaxError} {message: Invalid input 'C': expected whitespace, comment, DATABASE, DATABASES, DEFAULT, POPULATED, ALL, ROLES, USERS or show privilege scope (line 1, column 6 (offset: 5)) "SHOW CONSTRAINTS" ^} During handling of the above exception, another exception occurred: ValueError Traceback (most recent call last) [/tmp/ipykernel_3175/134052838.py](https://vscode-remote+dev-002dcontainer-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.vscode-resource.vscode-cdn.net/tmp/ipykernel_3175/134052838.py) in <module> 27 print(greeting) 28 ---> 29 graph = Neo4jGraph(url=url, username=username, password=password) 30 # QUERY = """ 31 # "MATCH (m:Movie)-[:IN_GENRE]->(:Genre {name:$genre}) [/usr/local/lib/python3.10/dist-packages/langchain_community/graphs/neo4j_graph.py](https://vscode-remote+dev-002dcontainer-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.vscode-resource.vscode-cdn.net/usr/local/lib/python3.10/dist-packages/langchain_community/graphs/neo4j_graph.py) in __init__(self, url, username, password, database, timeout, sanitize, refresh_schema) 217 if refresh_schema: 218 try: --> 219 self.refresh_schema() 220 except neo4j.exceptions.ClientError as e: 221 if e.code == "Neo.ClientError.Procedure.ProcedureNotFound": [/usr/local/lib/python3.10/dist-packages/langchain_community/graphs/neo4j_graph.py](https://vscode-remote+dev-002dcontainer-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.vscode-resource.vscode-cdn.net/usr/local/lib/python3.10/dist-packages/langchain_community/graphs/neo4j_graph.py) in refresh_schema(self) 281 # Get constraints & indexes 282 try: --> 283 constraint = self.query("SHOW CONSTRAINTS") 284 index = self.query("SHOW INDEXES YIELD *") 285 except ( [/usr/local/lib/python3.10/dist-packages/langchain_community/graphs/neo4j_graph.py](https://vscode-remote+dev-002dcontainer-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.vscode-resource.vscode-cdn.net/usr/local/lib/python3.10/dist-packages/langchain_community/graphs/neo4j_graph.py) in query(self, query, params) 250 return json_data 251 except CypherSyntaxError as e: --> 252 raise ValueError(f"Generated Cypher Statement is not valid\n{e}") 253 254 def refresh_schema(self) -> None: ValueError: Generated Cypher Statement is not valid {code: Neo.ClientError.Statement.SyntaxError} {message: Invalid input 'C': expected whitespace, comment, DATABASE, DATABASES, DEFAULT, POPULATED, ALL, ROLES, USERS or show privilege scope (line 1, column 6 (offset: 5)) "SHOW CONSTRAINTS" ^} ``` ### Description I am trying to load Neo4jGraph from langchain_community.graphs. Issue is simmilar to https://github.com/langchain-ai/langchain/discussions/17643 ; Hovewer in cited issue no solution was supplied . ### System Info langchain==0.1.15 langchain-community==0.0.32 langchain-core==0.1.41 langchain-experimental==0.0.57 langchain-openai==0.1.2 langchain-text-splitters==0.0.1 Python 3.10.12 Docker Ubuntu 20

Neo4j langchain integration error

https://api.github.com/repos/langchain-ai/langchain/issues/20262/comments

2024-04-10T08:17:00Z

2024-08-02T08:48:13Z

https://github.com/langchain-ai/langchain/issues/20262

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code server: ```python @chain def base_answer(source: Dict): prompt_b = ChatPromptTemplate.from_messages( [ ("system", system_prompt if source["from"] == "customer" else read_persist_var('system_prompt')), MessagesPlaceholder(variable_name="history"), ("human", "{input}") ] ) b_a = RunnableWithMessageHistory( prompt_b | llm(source["llm"]), get_message_history, input_messages_key="input", history_messages_key="history" ) | StrOutputParser() return b_a ``` client: ```Python base_answer = RemoteRunnable("http://localhost:2031/base_answer/") base_answer.stream({"input": "hi"}, config={"configurable": {"session_id": "d"}}) ``` ### Error Message and Stack Trace (if applicable) ValueError: Missing keys ['session_id'] in config['configurable'] Expected keys are I'session_id'].When using via .invoke() or stream(), pass in a config; e.g., chain. invoke(f'input': "too'}, {'configurable': {'session_id': "[your-value-here]"}}) ### Description   ### System Info python 3.9.18 langchian newest version ubuntu 20.04

Can't pass session id in

https://api.github.com/repos/langchain-ai/langchain/issues/20255/comments

2024-04-10T01:50:54Z

2024-07-18T16:08:59Z

https://github.com/langchain-ai/langchain/issues/20255

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python def api_add_docs(doc,collection_name): db = Chroma.from_documents(doc, persist_directory = PERSIST_DIRECTORY,collection_name=collection_name) db.persist() ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description I have a server up and running which takes document and collection name as an input to store in the Chroma db. The collection name is based on the user and there can be multiple users sending documents to same collection. As long as my API is up and running, all the documents coming from various sources are visible in the Chroma collections parquet file. However, when I stop the API, I see only the first document of first collection in the parquet file and rest all are erased. The code I have added, is just a representation of how I store and persist the db everytime I get a document through my API. ### System Info langchain==0.0.352 langchain-core==0.1.29 langchain-community==0.0.6 duckdb==0.10.2

Document Persist Temporarily

https://api.github.com/repos/langchain-ai/langchain/issues/20253/comments

2024-04-10T00:55:20Z

2024-07-18T16:08:54Z

https://github.com/langchain-ai/langchain/issues/20253

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code This code generates an error when invoked using chain.invoke from a prompt template. Removing any one of the stop sequences resolves the issue, but adding a stop sequence causes it. Content of stop sequences doesn't matter. ``` {python} llm = AzureOpenAI( api_version="2024-02-01", deployment_name = "my-test-deployment", max_tokens = 1500, temperature= 0.8, top_p= 1, model_kwargs= {"stop": ["<|im_end|>", "Student response:", "Grading:", "Test case", "Student Response:"]} ) ``` ### Error Message and Stack Trace (if applicable) BadRequestError: Error code: 400 - {'error': {'message': "'$.stop' is invalid. Please check the API reference: https://platform.openai.com/docs/api-reference.", 'type': 'invalid_request_error', 'param': None, 'code': None}} ### Description I'm trying to use langChain library to communicate with AzureOpenAI LLM, and using model kwargs to send stop sequences, but it seems that entering more than 4 stop sequences causes it to generate an error message. ### System Info System Information ------------------ > OS: Windows > OS Version: 10.0.22631 > Python Version: 3.12.2 (tags/v3.12.2:6abddd9, Feb 6 2024, 21:26:36) [MSC v.1937 64 bit (AMD64)] Package Information ------------------- > langchain_core: 0.1.40 > langchain: 0.1.14 > langchain_community: 0.0.31 > langsmith: 0.1.40 > langchain_openai: 0.1.1 > langchain_text_splitters: 0.0.1 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve

more than 4 stop sequences using model_kwargs for AzureOpenAI result in error

https://api.github.com/repos/langchain-ai/langchain/issues/20234/comments

2024-04-09T21:21:55Z

2024-04-10T14:42:48Z

https://github.com/langchain-ai/langchain/issues/20234

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ** ### Error Message and Stack Trace (if applicable) ** ### Description Please make a release of langchain-together that includes #19649 ### System Info **

Request for langchain-together release

https://api.github.com/repos/langchain-ai/langchain/issues/20217/comments

2024-04-09T17:34:41Z

2024-04-09T19:23:53Z

https://github.com/langchain-ai/langchain/issues/20217

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code model = genai.GenerativeModel(model_name='gemini-pro') #llm = OpenAI(temperature=0) mydb = SQLDatabase.from_uri('sqlite:///test.db') llm = {'model': model} toolkit=SQLDatabaseToolkit(db=mydb) agent_executer = create_sql_agent( llm=llm, toolkit=toolkit, handle_parsing_error=True, verbose=True ) ### Error Message and Stack Trace (if applicable) _No response_ ### Description Getting same issue , in my case issue is not getting resolved , is it because i am using gemini or different model from OpenAI, even i have followed all the recomendations from the chat , but still it is same, how to resolve the same issue using gemini pro model? genai.configure(api_key='my Api key') ### System Info import streamlit as st import pandas as pd import google.generativeai as genai from langchain_community.agent_toolkits import SQLDatabaseToolkit from langchain.agents.agent_types import AgentType from langchain.sql_database import SQLDatabase i have used above imoorts

#error: ValidationError: 1 validation error for SQLDatabaseToolkit

https://api.github.com/repos/langchain-ai/langchain/issues/20213/comments

2024-04-09T17:05:51Z

2024-05-22T11:13:44Z

https://github.com/langchain-ai/langchain/issues/20213

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code Code ``` from langchain.schema import Generation from langchain_core.output_parsers import JsonOutputParser print(JsonOutputParser().parse_result([Generation(text='{"key":"value\\')])) ``` Actual result: ```{}``` Expected result: not sure ### Error Message and Stack Trace (if applicable) _No response_ ### Description This problem occurred when I was streaming a JSON response containing new lines "\n". The chunk stopped right after the backslash, leading to an empty json after the parser. ### System Info langchain==0.1.14 langchain-core==0.1.40 plaform: linux Python 3.11.0rc1

JsonOutputParser returns empty json when text ends with "\"

https://api.github.com/repos/langchain-ai/langchain/issues/20204/comments

2024-04-09T12:33:52Z

2024-07-16T16:07:13Z

https://github.com/langchain-ai/langchain/issues/20204

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ``` python from langchain_openai import ChatOpenAI, OpenAIEmbeddings from langchain.chains import create_retrieval_chain from langchain.chains.combine_documents import create_stuff_documents_chain from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder from langchain_core.runnables.history import RunnableWithMessageHistory from langchain_core.runnables import ConfigurableFieldSpec, ConfigurableField from langchain_community.vectorstores.azuresearch import AzureSearch from langchain_community.chat_message_histories import RedisChatMessageHistory from langchain.chains import create_history_aware_retriever from typing import Any, Callable, Dict import os import re embeddings = OpenAIEmbeddings( openai_api_key=os.environ["OPENAI-API-KEY"], model="text-embedding-3-small" ) llm = ChatOpenAI( model_name="gpt-3.5-turbo-0125", temperature=0, openai_api_key=os.environ["OPENAI-API-KEY"] ) vector_store = AzureSearch( azure_search_endpoint=os.environ["VECTOR-STORE-ADDRESS"], azure_search_key=os.environ["VECTOR-STORE-ADMIN-KEY"], index_name="products", embedding_function=embeddings.embed_query, ) def _is_valid_identifier(value: str) -> bool: """ Validate an identifier such as user_id or conversation_id. Args: - value (str): The identifier to be validated. Returns: - bool: True if the identifier is valid, False otherwise. """ valid_characters = re.compile(r"^[a-zA-Z0-9-_]+$") return bool(valid_characters.match(value)) def create_session_factory() -> Callable[[str, str], RedisChatMessageHistory]: """ Create a session factory for Redis based chat history storage. Returns: - Callable[[str, str], RedisChatMessageHistory]: A function to get chat history based on user_id and conversation_id. """ def get_chat_history(user_id: str, conversation_id: str) -> RedisChatMessageHistory: if not _is_valid_identifier(user_id): raise ValueError( f"User ID {user_id} is not in a valid format. " "User ID must only contain alphanumeric characters, " "hyphens, and underscores." ) if not _is_valid_identifier(conversation_id): raise ValueError( f"Conversation ID {conversation_id} is not in a valid format. " "Conversation ID must only contain alphanumeric characters, " "hyphens, and underscores." ) return RedisChatMessageHistory( session_id=f"user:{user_id}:conversation:{conversation_id}", url="redis://172.20.0.3:6379", key_prefix="qna-rag" ) return get_chat_history contextualize_q_system_prompt = """Given a chat history and the latest user question \ which might reference context in the chat history, formulate a standalone question \ which can be understood without the chat history. Do NOT answer the question, \ just reformulate it if needed and otherwise return it as is.""" contextualize_q_prompt = ChatPromptTemplate.from_messages( [ ("system", contextualize_q_system_prompt), MessagesPlaceholder("chat_history"), ("human", "{input}"), ] ) history_aware_retriever = create_history_aware_retriever( llm, vector_store.as_retriever( search_kwargs={"k": 2} ).configurable_fields( search_kwargs=ConfigurableField( id="search_kwargs_retriever", name="Search Kwargs", description="The search kwargs to use", ) ), contextualize_q_prompt ) qa_system_prompt = """Answer the user's questions based on the below context. If the context doesn't contain any relevant information to the question, don't make something up and just say "I don't know": <context> {context} </context>""" qa_prompt = ChatPromptTemplate.from_messages( [ ("system", qa_system_prompt), MessagesPlaceholder("chat_history"), ("human", "{input}"), ] ) question_answer_chain = create_stuff_documents_chain(llm, qa_prompt) rag_chain = create_retrieval_chain(history_aware_retriever, question_answer_chain) conversational_rag_chain = RunnableWithMessageHistory( rag_chain, create_session_factory(), input_messages_key="input", history_messages_key="chat_history", history_factory_config=[ ConfigurableFieldSpec( id="user_id", annotation=str, name="User ID", description="Unique identifier for the user.", default="", is_shared=True, ), ConfigurableFieldSpec( id="conversation_id", annotation=str, name="Conversation ID", description="Unique identifier for the conversation.", default="", is_shared=True, ), ], output_messages_key="answer" ) ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description I do not know how to add sources to the rag chain up above, I tried multiple solutions like using load_qa_with_sources chain but that doesn't seem to work. I also tried [this](https://python.langchain.com/docs/use_cases/question_answering/sources/#adding-sources) but am unable to figure out how to add it into my conversational_rag_chain above. All I want to do is return sources along with the answer for the conversational_rag_chain. I did try out what was recommended in this [discussion](https://github.com/langchain-ai/langchain/discussions/16582) still not able to make it work. No help on discussions, kapa_ai_bot or dosubot or the community, hence the issue here ### System Info ``` System Information ------------------ > OS: Linux > OS Version: #1 SMP Wed Mar 2 00:30:59 UTC 2022 > Python Version: 3.11.8 (main, Feb 7 2024, 22:38:59) [GCC 12.2.0] Package Information ------------------- > langchain_core: 0.1.35 > langchain: 0.1.13 > langchain_community: 0.0.29 > langsmith: 0.1.31 > langchain_cli: 0.0.21 > langchain_openai: 0.0.8 > langchain_text_splitters: 0.0.1 > langchainhub: 0.1.15 > langserve: 0.0.51 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph ```

Can't add sources to RunnableWithMessageHistory

https://api.github.com/repos/langchain-ai/langchain/issues/20203/comments

2024-04-09T11:37:14Z

2024-04-23T06:02:53Z

https://github.com/langchain-ai/langchain/issues/20203

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python from langchain_community.document_loaders import GCSDirectoryLoader loader = GCSDirectoryLoader(project_name="aist", bucket="testing-hwc") loader.load() # Error occurs here ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description Users reading in large number of docs using `GCSDirectoryLoader` may experience exceptions. They will break the function as the exception is not caught in try-exception block. Previous PR #19591 has attempted to solve this issue but there's an error in the code. Issue since langchain-community==0.0.31 PR #20005 will resolve this issue, please merge. ### System Info Issue since langchain-community==0.0.31

Exception not caught in try-exception block in GCSDirectoryLoader

https://api.github.com/repos/langchain-ai/langchain/issues/20198/comments

2024-04-09T09:01:22Z

2024-07-16T16:07:08Z

https://github.com/langchain-ai/langchain/issues/20198

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ``` %pip install --upgrade --quiet langchain langchain-openai from langchain_openai import ChatOpenAI llm = ChatOpenAI(api_key="sk-ssot-xxx-xxx-xxx", base_url="https://openai-proxy.xxxx.is/v1") llm.invoke("how can langsmith help with testing?") ``` ### Error Message and Stack Trace (if applicable) --------------------------------------------------------------------------- AttributeError Traceback (most recent call last) Cell In[62], line 1 ----> 1 llm.invoke("how can langsmith help with testing?") File /opt/app-root/lib64/python3.8/site-packages/langchain_core/language_models/chat_models.py:173, in invoke(self, input, config, stop, **kwargs) 165 async def ainvoke( 166 self, 167 input: LanguageModelInput, (...) 171 **kwargs: Any, 172 ) -> BaseMessage: --> 173 config = ensure_config(config) 174 llm_result = await self.agenerate_prompt( 175 [self._convert_input(input)], 176 stop=stop, (...) 181 **kwargs, 182 ) 183 return cast(ChatGeneration, llm_result.generations[0][0]).message File /opt/app-root/lib64/python3.8/site-packages/langchain_core/language_models/chat_models.py:571, in generate_prompt(self, prompts, stop, callbacks, **kwargs) 563 prompt_messages = [p.to_messages() for p in prompts] 564 return await self.agenerate( 565 prompt_messages, stop=stop, callbacks=callbacks, **kwargs 566 ) 568 def _generate_with_cache( 569 self, 570 messages: List[BaseMessage], --> 571 stop: Optional[List[str]] = None, 572 run_manager: Optional[CallbackManagerForLLMRun] = None, 573 **kwargs: Any, 574 ) -> ChatResult: 575 if isinstance(self.cache, BaseCache): 576 llm_cache = self.cache File /opt/app-root/lib64/python3.8/site-packages/langchain_core/language_models/chat_models.py:403, in generate(self, messages, stop, callbacks, tags, metadata, run_name, **kwargs) 401 results = [] 402 for i, m in enumerate(messages): --> 403 try: 404 results.append( 405 self._generate_with_cache( 406 m, (...) 410 ) 411 ) 412 except BaseException as e: File /opt/app-root/lib64/python3.8/site-packages/langchain_core/callbacks/manager.py:1443, in configure(cls, inheritable_callbacks, local_callbacks, verbose, inheritable_tags, local_tags, inheritable_metadata, local_metadata) 1418 @classmethod 1419 def configure( 1420 cls, (...) 1427 local_metadata: Optional[Dict[str, Any]] = None, 1428 ) -> CallbackManager: 1429 """Configure the callback manager. 1430 1431 Args: 1432 inheritable_callbacks (Optional[Callbacks], optional): The inheritable 1433 callbacks. Defaults to None. 1434 local_callbacks (Optional[Callbacks], optional): The local callbacks. 1435 Defaults to None. 1436 verbose (bool, optional): Whether to enable verbose mode. Defaults to False. 1437 inheritable_tags (Optional[List[str]], optional): The inheritable tags. 1438 Defaults to None. 1439 local_tags (Optional[List[str]], optional): The local tags. 1440 Defaults to None. 1441 inheritable_metadata (Optional[Dict[str, Any]], optional): The inheritable 1442 metadata. Defaults to None. -> 1443 local_metadata (Optional[Dict[str, Any]], optional): The local metadata. 1444 Defaults to None. 1445 1446 Returns: 1447 CallbackManager: The configured callback manager. 1448 """ 1449 return _configure( 1450 cls, 1451 inheritable_callbacks, (...) 1457 local_metadata, 1458 ) File /opt/app-root/lib64/python3.8/site-packages/langchain_core/callbacks/manager.py:1940, in _configure(callback_manager_cls, inheritable_callbacks, local_callbacks, verbose, inheritable_tags, local_tags, inheritable_metadata, local_metadata) 1926 callback_manager = callback_manager_cls( 1927 handlers=inheritable_callbacks_.copy(), 1928 inheritable_handlers=inheritable_callbacks_.copy(), 1929 parent_run_id=parent_run_id, 1930 ) 1931 else: 1932 callback_manager = callback_manager_cls( 1933 handlers=inheritable_callbacks.handlers.copy(), 1934 inheritable_handlers=inheritable_callbacks.inheritable_handlers.copy(), 1935 parent_run_id=inheritable_callbacks.parent_run_id, 1936 tags=inheritable_callbacks.tags.copy(), 1937 inheritable_tags=inheritable_callbacks.inheritable_tags.copy(), 1938 metadata=inheritable_callbacks.metadata.copy(), 1939 inheritable_metadata=inheritable_callbacks.inheritable_metadata.copy(), -> 1940 ) 1941 local_handlers_ = ( 1942 local_callbacks 1943 if isinstance(local_callbacks, list) 1944 else (local_callbacks.handlers if local_callbacks else []) 1945 ) 1946 for handler in local_handlers_: AttributeError: 'NoneType' object has no attribute 'get' ### Description I'm testing langchain and our OpenAI is behind our proxy. It is work for the below code: ``` %pip install --upgrade --quiet openai from openai import OpenAI client = OpenAI(api_key="sk-ssot-xxx-xxx-xxx", base_url="https://openai-proxy.xxxx.is/v1") completion = client.chat.completions.create( model="gpt-3.5-turbo", messages=[ {"role": "system", "content": "You are a poetic assistant, skilled in explaining complex programming concepts with creative flair."}, {"role": "user", "content": "Compose a poem that explains the concept of recursion in programming."} ] ) print(completion.choices[0].message) ``` but under the same simple setup, Langchain is throwing me the above error. code to produce the error: ``` %pip install --upgrade --quiet langchain langchain-openai from langchain_openai import ChatOpenAI llm = ChatOpenAI(api_key="sk-ssot-xxx-xxx-xxx", base_url="https://openai-proxy.xxxx.is/v1") llm.invoke("how can langsmith help with testing?") ``` ### System Info langchain==0.1.14 langchain-community==0.0.31 langchain-core==0.1.40 langchain-openai==0.1.1 langchain-text-splitters==0.0.1 Linux 5.15.0-83-generic x86_64 NAME="CentOS Linux" VERSION="7 (Core)" ID="centos" ID_LIKE="rhel fedora" VERSION_ID="7" PRETTY_NAME="CentOS Linux 7 (Core)" ANSI_COLOR="0;31" CPE_NAME="cpe:/o:centos:centos:7" HOME_URL="https://www.centos.org/" BUG_REPORT_URL="https://bugs.centos.org/" CENTOS_MANTISBT_PROJECT="CentOS-7" CENTOS_MANTISBT_PROJECT_VERSION="7" REDHAT_SUPPORT_PRODUCT="centos" REDHAT_SUPPORT_PRODUCT_VERSION="7" Python 3.8.6

Cannot connect to OpenAI by providing api_key and base_url

https://api.github.com/repos/langchain-ai/langchain/issues/20195/comments

2024-04-09T07:40:06Z

2024-07-16T16:07:03Z

https://github.com/langchain-ai/langchain/issues/20195

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code import os; from langchain.llms.openai import OpenAI from openai import AzureOpenAI from dotenv import load_dotenv from langchain.chains.api.base import LLMChain from langchain.chains.api.base import APIChain from langchain.chains.api import open_meteo_docs from langchain_core.prompts import PromptTemplate load_dotenv() client = AzureOpenAI( api_key = os.getenv("AZURE_OPENAI_API_KEY"), api_version = "2024-02-15-preview", azure_endpoint = os.getenv("AZURE_OPENAI_ENDPOINT") ) prompt = PromptTemplate( input_variables=["api_url"], template="""Act as a technical writer. Write detailed documentation for the API that exists at {api_url}. Only detail the request, do not describe the response. Do not include any parameters not in the sample endpoint.""" ) chain = LLMChain( llm=client, verbose=True, prompt=prompt ) url = "https://pitchfork.com/api/v2/search/?genre=experimental&genre=global&genre=jazz&genre=metal&genre=pop&genre=rap&genre=rock&types=reviews&sort=publishdate%20desc%2Cposition%20asc&size=5&start=0&rating_from=0.0" response = chain.run(url) print(response) ### Error Message and Stack Trace (if applicable) super().__init__(**kwargs) File "pydantic\main.py", line 341, in pydantic.main.BaseModel.__init__ pydantic.error_wrappers.ValidationError: 2 validation errors for LLMChain llm instance of Runnable expected (type=type_error.arbitrary_type; expected_arbitrary_type=Runnable) llm instance of Runnable expected (type=type_error.arbitrary_type; expected_arbitrary_type=Runnable) ### Description I am trying to call an external api. Here i am using langchain version==0.1.0 pydantic==1.10.8 and openai==1.7.0 I am using azureopenai instead openai.. and whenever i want to use the LLMChain , i got this 2 validation error. I am confused that whether this is because of any version issue or the implementation of the code. ### System Info Python 3.10.11 langchain version==0.1.0 pydantic==1.10.8 openai==1.7.0 openapi-schema-pydantic==1.2.4

pydantic.error_wrappers.ValidationError: 2 validation errors for LLMChain

https://api.github.com/repos/langchain-ai/langchain/issues/20192/comments

2024-04-09T06:53:07Z

2024-07-19T16:08:06Z

https://github.com/langchain-ai/langchain/issues/20192

### Privileged issue - [X] I am a LangChain maintainer, or was asked directly by a LangChain maintainer to create an issue here. ### Issue Content Only some function/tool-calling models have a bind_tools method. Others have a bind_functions method, and others still don't have any special bind methods. This is in part due to different api parameter names, in part due to differences in model parallel tool/function calling abilities, in part because we were waiting to see if function/tool-calling would become widespread. It seems like it has become widespread and the name we're converging on for parallel function/tool calling (which is the more common and more generic interface than single function calling) is just "tool calling". So proposing that we standardize the bind_tools interface and have all capable models implement it. Standard interface would be something like ```python class BaseLanguageModel(...): def bind_tools( self, tools: Sequence[Union[Dict[str, Any], Type[BaseModel], Callable, BaseTool]], **kwargs: Any, ) -> Runnable[LanguageModelInput, BaseMessage]: ... ```

RFC: add `bind_tools` to BaseLanguageModel

https://api.github.com/repos/langchain-ai/langchain/issues/20178/comments

2024-04-08T21:27:27Z

2024-08-08T00:31:05Z

https://github.com/langchain-ai/langchain/issues/20178

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code # Full code at https://github.com/digillia/Digillia-Colab/blob/main/tools/langchain.ipynb from langchain.chains import create_history_aware_retriever, create_retrieval_chain from langchain.chains.combine_documents import create_stuff_documents_chain condense_prompt = ChatPromptTemplate.from_messages([ SystemMessage(content=SYSTEM_PROMPT), HumanMessagePromptTemplate.from_template(CONDENSE_PROMPT), ]) history_aware_retriever = create_history_aware_retriever(llm, retriever, condense_prompt) context_prompt = ChatPromptTemplate.from_messages([ SystemMessagePromptTemplate.from_template(SYSTEM_PROMPT + '\n' + CONTEXT_PROMPT), HumanMessagePromptTemplate.from_template('{input}') ]) question_answer_chain = create_stuff_documents_chain(llm, context_prompt) rag_chain = create_retrieval_chain(history_aware_retriever, question_answer_chain) ### Error Message and Stack Trace (if applicable) This more a design issue rather than a bug. ### Description IMHO `create_history_aware_retriever` should return both: - the rephrased question considering the chat history, and - the documents retrieved based on the rephrased question. See https://github.com/langchain-ai/langchain/discussions/20128 ### System Info System Information OS: Darwin OS Version: Darwin Kernel Version 23.4.0: Fri Mar 15 00:12:41 PDT 2024; root:xnu-10063.101.17~1/RELEASE_ARM64_T8103 Python Version: 3.9.6 (default, Feb 3 2024, 15:58:27) [Clang 15.0.0 (clang-1500.3.9.4)] Package Information langchain_core: 0.1.40 langchain: 0.1.14 langchain_community: 0.0.31 langsmith: 0.1.40 langchain_openai: 0.0.2.post1 langchain_text_splitters: 0.0.1 Packages not installed (Not Necessarily a Problem) The following packages were not found: langgraph langserve

Probable design issue with create_retrieval_chain from create_history_aware_retriever

https://api.github.com/repos/langchain-ai/langchain/issues/20156/comments

2024-04-08T12:21:44Z

2024-07-15T16:07:07Z

https://github.com/langchain-ai/langchain/issues/20156

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code NA ### Error Message and Stack Trace (if applicable) NA ### Description The PDF is actually in Hindi Format and when I am reading it is printing in this language. ``` "cht “kks/ku ds fy, Fkhje ¼2 xzke@fdxzk cht½ dk iz;ksx chtksa dks laØfer\ngksus ls cpkus ds fy, djrs gSA DyksjksFkSyksfuy 0-2 izfr”kr ;k eSadkstsc 0-2\nizfr”kr dh nj ls fNM+dko djds izHkkoh jksx fu;a=.k fd;k tk ldrk gSA\nNksVh iÙkh chekjh ¼fyfVy yhQ½ ekbdksIykLek ls gksrh gS tks yhQ gkWij ds\n¼lhfLVl fglheksul½ dkilkbfVl }kjk QSyrh gSA izHkkfor ikS/kk vkdkj esa\nNksVk vkSj ifÙk;ka iSjk NksVh] vfodflr uiqald Qwy tks fd dksbZ Qy dk\n/kkj.k ugh djrhA jksdFkke ds mik; esa fiNsrh Qlyksa dks bdV~Bk djds\nmudks tyk nsuk] “kq:vkrh nkSj ds izHkkfor ikS/kksa dks [ksr ls fudkyuk] rkfd\nfcekjh dk iquZforj.k u gks ik;s vkSj igyh “kq:vkrh y{k.k ds fn[kkbZ nsus ij\neSykfFk;ku 0-2 izfr”kr dh nj fNM+dko djus ls jksx dks fu;af=r fd;k tk\nldrk gSA\niz'u& 4 % mÙkj izns'k ns'k esa fepZ mRiknd {ks=ksa esa ls ,d gS vkSj cgq/kk ge\nfepZ ds iÙkh ds /kCcsnkj jksx dk lkeuk djrs gS] bldk izca/ku dSls\ndjas\\\nmÙkj % iÙkh ds /kCcsnkj jksx tSls ,UFkzksDukst] vYVjusfj;k ldksZLiksjk yhQ LikV vkSj\nyhQ dyZ dkEiysDl bR;kfn chekfj;ka gSa tks fepZ esa yxrh gS\n,UFkSDukst% ;g dksykWbVksVªkbde dSIlkbZ dod ds }kjk QSyrk gSA blds ds\ny{k.k “kq:vkrh nkSj esa u;h Vgfu;ksa ij ¼usØksfll½ Åijh fljs ls uhps dh\nrjQ gfj;kyh foghu fn[kk;h iM+rs gS blfy, bls MkbZcSd Hkh dgrs gSaA Qyksa\nij “kq:vkrh nkSj esa gYds Hkwjs jax ds /kwlj CykbV dh rjg fn[kk;h nsrh gSA\nLkjdksLiksjk yhQ LikV % bl fcekjh ds y{k.k xqykch jax ds xksy /kCcs\nftuds e/; esa Hkwjs jax dk dsUnz gksrk gS ,oa vYVjusfj;k ls izHkkfor ikS/kksa esa\nxgjs Hkwjs jax ds vfuf;er /kCcs ik;s tkrs gS fdUrq nksuksa CykbV ds y{k.k nsj esa\nfn[kk;h iM+rs gSaA\n,aFkszDukst] ldksZLiksjk vkSj vYVjusfj;k yhQ LikV lHkh ds fu;a=.k fy,\nDyksjksFkSyksfuy ¼0-2 izfr”kr½ ,oa eSadkstsc ¼0-2 izfr”kr½ ,d ds ckn 15 fnuksa\nds vUrjky ij fNM+dko djuk pkfg,A\nyhQ dyZ % fepZ esa iÙkh dk xqjpk jksx dbZ dkj.kkas ls gksrk gS tSls fo’kk.kq\neD[kh vkSj fFkzIl ds dkj.k y{k.k eq[;r;k vfodflr >qylh gq;h] [kqjnjh ,oa\neksVh] uhps dh rjQ eqM+h ,oa xqPNs dk vkdkj /kkj.k dj ysrh gSA chekjh ds\nizca/ku ds fy, ikS/ks dh tM+ksa dks bfeMkDyksfizM ¼4 feyh@yhVj ikuh½ esa ,d\n?kaVs ds fy, Mqcksuk pkfg,] ikS/k dks dhVjks/kh tky ds vUnj mxkuk pkfg,]\nvojks/kh Qlyksa ¼eDdk] cktjk] Tokj bR;kfn½ dks yxkuk pkfg, ftlls dhVksa\nds MM+us esa ck/kk mRiUu gksrh gS vkSj le;&le; ij vUrjokgh flLVfed\ndhVuk”kd dk iz;ksx djuk pkfg,A\n44\n" ``` ### System Info pip == 24.0 python == 3.10.10 Cuda == 12.1

I'm using langchain document_loaders to read a PDF, but it's in Hindi, and I'm having trouble understanding it.

https://api.github.com/repos/langchain-ai/langchain/issues/20153/comments

2024-04-08T11:47:07Z

2024-04-08T15:39:45Z

https://github.com/langchain-ai/langchain/issues/20153

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code the following code: ``` metaqa_path = pathlib.Path(__file__).parent.absolute() / 'mydata' / 'metaQA' metaqa_rdf = metaqa_path / 'knowledge_graph.rdf' metaqa_ttl = metaqa_path / 'kb_wo_parse.ttl' graph = RdfGraph( source_file=str(metaqa_ttl), standard="rdf", ) chain = GraphSparqlQAChain.from_llm( ChatOpenAI(temperature=0), graph=graph, verbose=True ) print(chain.run(r'which movies starred the actor Gert Fröbe')) ``` ### Error Message and Stack Trace (if applicable) line 1200, in parse_string raise exc.with_traceback(None) pyparsing.exceptions.ParseException: Expected {SelectQuery | ConstructQuery | DescribeQuery | AskQuery}, found '`' (at char 0), (line:1, col:1) ### Description - It should be eliminitaing the ``` (triple quotes) generated by the LLM, that cannot be parsed by RDFLIB. - Should generate ``` PREFIX metaQAKB: <http://metaQAKB.org/> SELECT ?movie WHERE { ?movie metaQAKB:starred_actors "Gert Fröbe" . } ``` Instead of ``` ``` PREFIX metaQAKB: <http://metaQAKB.org/> SELECT ?movie WHERE { ?movie metaQAKB:starred_actors "Gert Fröbe" . } ``` ``` ### System Info langchain-0.1.14 rdflib-7.0.0 langchain-openai-0.1.1

Parsing of generated SPARQL via langchain rdflib always gives "pyparsing.exceptions.ParseException"

https://api.github.com/repos/langchain-ai/langchain/issues/20150/comments

2024-04-08T11:05:26Z

2024-05-25T20:44:17Z

https://github.com/langchain-ai/langchain/issues/20150

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code nil ### Error Message and Stack Trace (if applicable) _No response_ ### Description is there any way to get the Openai API key usage and balance through langchain in python code? ### System Info python: 3.11 langchain: latest

how to get openai usage through code

https://api.github.com/repos/langchain-ai/langchain/issues/20139/comments

2024-04-08T07:45:21Z

2024-04-08T15:41:21Z

https://github.com/langchain-ai/langchain/issues/20139

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ``` from langchain.agents import AgentExecutor from langchain.agents.output_parsers.openai_tools import \ OpenAIToolsAgentOutputParser from langchain.memory import ConversationBufferMemory from langchain_community.tools import DuckDuckGoSearchRun from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder from langchain_openai import ChatOpenAI systm = "As a research agent named Geppetto, you are adept at utilizing diverse analytical tools that you have access to in order to address inquiries effectively. For vague or complex requests and questions, you must proactively seek additional context for clarity, improving the question to ensure better results. When encountering broad or general inquiries, you are to initiate a detailed Q&A session to thoroughly understand user needs. Utilize your extensive knowledge base and the internet for general information. Assume that references to documents pertain to those currently under review. Always review your response and improve." llm = ChatOpenAI(temperature=0, model="gpt-4-0125-preview",streaming=True) prompt = ChatPromptTemplate.from_messages( [ ( "system", systm, ), ("user", "{input}"), MessagesPlaceholder(variable_name="agent_scratchpad"), MessagesPlaceholder(variable_name="chat_history"), ] ) memory = ConversationBufferMemory(memory_key="chat_history",return_messages=True) # Construct the JSON agent fileTools = [DuckDuckGoSearchRun()] llm_with_tools = llm.bind_tools(fileTools) from langchain.agents.format_scratchpad.openai_tools import \ format_to_openai_tool_messages agent = ( { "input": lambda x: x["input"], "agent_scratchpad": lambda x: format_to_openai_tool_messages( x["intermediate_steps"] ), "chat_history": lambda x: x["chat_history"] } | prompt | llm_with_tools | OpenAIToolsAgentOutputParser() ) agent_executor = AgentExecutor(agent=agent, tools=fileTools, verbose=False,memory=memory) while True: print() req = input("What is your requesat/question? ") print() ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description As you can see from below, the chat conversation is not tracked appropriately with memory. What is your requesat/question? **How many people live in canada?** As of April 6, 2024, the population of Canada is 39,029,568. What is your requesat/question? **what is their national anthem called?** A_**s of July 1, 2023, Canada's population was estimated to be 40,097,761.**_ What is your requesat/question? **what is the population of Indonesia** **_Canada's national anthem is called "O Canada."_** What is your requesat/question? what have we discussed? _**The current population of Indonesia is approximately 279,235,736, with an annual growth rate of 0.798%.**_ What is your requesat/question? ### System Info langchain==0.1.13 langchain-anthropic==0.1.4 langchain-community==0.0.29 langchain-core==0.1.33 langchain-groq==0.0.1 langchain-openai==0.1.1 langchain-text-splitters==0.0.1 langchainhub==0.1.14 Python 3.11.7 Platform MAC

AgentExecutor: Memory: Some answers are responses to prior questions. conversation not tracked appropriately

https://api.github.com/repos/langchain-ai/langchain/issues/20125/comments

2024-04-07T12:53:21Z

2024-07-11T03:23:32Z

https://github.com/langchain-ai/langchain/issues/20125

### Checklist - [X] I added a very descriptive title to this issue. - [X] I included a link to the documentation page I am referring to (if applicable). ### Issue with current documentation: **URL to Documentation** https://python.langchain.com/docs/modules/memory/agent_with_memory/ **Warning when running the code:** /Users/randolphhill/govbotics/development-Indonesia/chatdoc/.venv/lib/python3.11/site-packages/langchain_core/_api/deprecation.py:117: LangChainDeprecationWarning: The class `langchain.agents.mrkl.base.ZeroShotAgent` was deprecated in langchain 0.1.0 and will be removed in 0.2.0. Use create_react_agent instead. **May I suggest you have someone assigned to clean up the documentation or hire a contractor to fix. I have found these types of mistakes. In many placed. This time I decided to submit a bug** ### Idea or request for content: Need up update the documentation. **https://python.langchain.com/docs/modules/memory/agent_with_memory/**

DOC: Memory in Agent Documentation needs to be updated.

https://api.github.com/repos/langchain-ai/langchain/issues/20122/comments

2024-04-07T12:03:03Z

2024-04-17T20:35:16Z

https://github.com/langchain-ai/langchain/issues/20122

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code if __name__ == '__main__': input = 'where is my dog?' #create embedding function by using model of 'textembedding-gecko@003' vertexai_embedding_003 = VertexAIEmbeddings(model_name='textembedding-gecko@003') # init a pinecone vectorstore with vertex ai embedding pc = Pinecone(api_key=os.getenv("PINECONE_API_KEY"), environment='us-central1-gcp') vector_store = PineconeVectorStore(index_name='embedding-test', embedding=vertexai_embedding_003) # create a test document doc = Document( page_content=input, metadata={'category': 'pet'} ) # save in the index vector_store.add_documents([doc]) # similarity search from data we inserted before print(vector_store.similarity_search_with_score(input)) ### Error Message and Stack Trace (if applicable) Screenshot of different vectors by embedding the same input('where is my dog?') Embedding result when doing insertion <img width="891" alt="Screenshot 2024-04-07 at 16 51 36" src="https://github.com/langchain-ai/langchain/assets/62688587/356117a3-5626-4e47-930e-bee7cd42fdb0"> Embedding result when doing query <img width="999" alt="Screenshot 2024-04-07 at 16 51 19" src="https://github.com/langchain-ai/langchain/assets/62688587/bc70532c-83b3-46aa-8aad-9a8285849996"> _No response_ ### Description Hello Langchain team, I found the embedding issue between adding embedding in pinecone and do similarity_search_with_score from pinecone by using the model of 'textembedding-gecko@003' of google vertex ai. It only happen on 'textembedding-gecko@003', **for 'textembedding-gecko@001' works fine** How to reproduce 1, adding input string by using vector_store.add_documents([doc]), before it does insertion, the code will calculate the vectors by 'textembedding-gecko@003'. And then it will store the vectors and metadata into vectorstore. 2, And if we search the exactly same string by using function of 'similarity_search_with_score', our expectation score should be 1, because the input query is the same. But actually, it return '0.79' due to the wrong embedding result After I debug the code and I found there is issue of embedding ways between stage of adding document and stage of searching document. here is the sreenshot  We can see adding documents and query documents passed the different 'embedding_task_type' which is the reason of giving the different embedding result by passing the same input And meanwhile parameter of 'embedding_task_type' is hardcode for these to functions, user is not able to customized it. Here is the doc of explanation of google https://cloud.google.com/python/docs/reference/aiplatform/latest/vertexai.language_models.TextEmbeddingInput. Conclusion, if devs follow the documents of langchian to inert and query by using 'textembedding-gecko@003', it is very easy to meet the this issue ### System Info langchain==0.1.14 langchain_google_vertexai==0.1.2 langchain-pinecone==0.0.3

Intergation issue between langchain-pinecone and google vertex AI textembedding-gecko@003

https://api.github.com/repos/langchain-ai/langchain/issues/20118/comments

2024-04-07T08:43:23Z

2024-07-23T16:08:16Z

https://github.com/langchain-ai/langchain/issues/20118

### Checklist - [X] I added a very descriptive title to this issue. - [X] I included a link to the documentation page I am referring to (if applicable). ### Issue with current documentation: TypeError: 'FAISS' object is not callable Traceback: File "C:\Users\Jashu\AppData\Local\Programs\Python\Python311\Lib\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 542, in _run_script exec(code, module.__dict__) File "C:\Medi_LLM\Llama-2-7B-Chat-GGML\app.py", line 34, in <module> retriever = vector_store(search_kwargs=["k: 2"]), ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ### Idea or request for content: _No response_

DOC: <Please write a comprehensive title after the 'DOC: ' prefix>TypeError: 'FAISS' object is not callable Traceback: File "C:\Users\Jashu\AppData\Local\Programs\Python\Python311\Lib\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 542, in _run_script exec(code, module.__dict__) File "C:\Medi_LLM\Llama-2-7B-Chat-GGML\app.py", line 34, in <module> retriever = vector_store(search_kwargs=["k: 2"]), ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

https://api.github.com/repos/langchain-ai/langchain/issues/20112/comments

2024-04-06T19:24:28Z

2024-04-08T15:40:37Z

https://github.com/langchain-ai/langchain/issues/20112

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [ ] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code getting error at this part: ``` prompt.format( question="Give a JSON of all the reward categories corresponding points and required details for each category from the given context", context=tst_context) ``` ### Error Message and Stack Trace (if applicable) `KeyError: "'source'"` ### Description [Examples.txt](https://github.com/langchain-ai/langchain/files/14894264/Examples.txt) I am trying to scrap a website using WebBaseLoader and generate a context then ask LLM some questions on the generated context and instructing it to give the output as JSON. I am using the FewShotPromptTemplate; Examples: _I have attached the file_( Giving a preview) ``` examples=[ { "context": context1, "question": "Give a JSON of all the reward categories corresponding points and required details for each category from the given context", "answer": """ {{"credit_cards": [ { "card_name": "SBI Prime Credit Card", "reward_categories": [ { "category": "Birthday Spends", "points_per_transaction": 20, "details": "20 reward points for every Rs. 100 spent on your birthday*", "capped_points": "Reward points earned on birthday spends (one day before, on, and one day after) are capped at 2,000 reward points per calendar year." }, { "category": "Dining", "points_per_transaction": 10, "details": "10 reward points for every Rs. 100 spent on dining" }, { "category": "Groceries", "points_per_transaction": 10, "details": "10 reward points for every Rs. 100 spent on groceries" }, { "category": "Departmental Stores", "points_per_transaction": 10, "details": "10 reward points for every Rs. 100 spent on departmental stores" }, { "category": "Movies", "points_per_transaction": 10, "details": "10 reward points for every Rs. 100 spent on movies" }, { "category": "All Other Retail Purchases (Except Fuel)", "points_per_transaction": 2, "details": "2 reward points for every Rs. 100 spent on all other retail purchases, except for fuel" } ] } ] }}""" } ``` Example prompt: ``` example_prompt = PromptTemplate( input_variables=["context", "question","answer"], template = """Answer the following question based only on the provided context: <context> {context} </context> Question: {question}\n Answer:{answer}""" ) ``` Promt: ``` prompt = FewShotPromptTemplate( examples=examples, example_prompt=example_prompt, suffix="""Answer the following question based only on the provided context: \n <context> {context} </context>\n Question: {input}\n Answer: """, input_variables=["input","context"], example_separator="\n\n" ) ``` ``` prompt.format( input="Give a JSON of all the reward categories corresponding points and required details for each category from the given context", context=tst_context) ``` ### System Info On Google colab !pip install -q langchain

while using FewShotPromptTemplate, getting KeyError: "'source'";But in my template there is no such key as 'source'

https://api.github.com/repos/langchain-ai/langchain/issues/20108/comments

2024-04-06T17:29:45Z

2024-07-13T16:06:11Z

https://github.com/langchain-ai/langchain/issues/20108

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code llm = ChatOpenAI(model_name="gpt-3.5-turbo", temperature = 0) llm_with_tools = llm.bind_tools( tools=tools, tool_choice="auto", ) tools = [ { "type": "function", "function": { "name": "answer_queries_with_citations", "description": "useful for answering questions about article or document content.", "parameters": {}, "required": [], } }, { "type": "function", "function": { "name": "miscellaneous_question_answerer", "description": "useful for answering miscellaneous questions.", "parameters": {}, "required": [], } }, { "type": "function", "function": { "name": "summarize_document", "description": "useful when very specifically asked for a summary of the document.", "parameters": {}, "required": [], } } ] ### Error Message and Stack Trace (if applicable) [2024-04-06 09:55:56,359] ERROR in app: Exception on /docquery [POST] Traceback (most recent call last): File "/opt/render/project/src/.venv/lib/python3.11/site-packages/flask/app.py", line 1463, in wsgi_app response = self.full_dispatch_request() ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/render/project/src/.venv/lib/python3.11/site-packages/flask/app.py", line 872, in full_dispatch_request rv = self.handle_user_exception(e) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/render/project/src/.venv/lib/python3.11/site-packages/flask_cors/extension.py", line 176, in wrapped_function return cors_after_request(app.make_response(f(*args, **kwargs))) ^^^^^^^^^^^^^^^^^^ File "/opt/render/project/src/.venv/lib/python3.11/site-packages/flask/app.py", line 870, in full_dispatch_request rv = self.dispatch_request() ^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/render/project/src/.venv/lib/python3.11/site-packages/flask/app.py", line 855, in dispatch_request return self.ensure_sync(self.view_functions[rule.endpoint])(**view_args) # type: ignore[no-any-return] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/render/project/src/.venv/lib/python3.11/site-packages/flask_httpauth.py", line 174, in decorated return self.ensure_sync(f)(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/render/project/src/app.py", line 108, in test llm_with_tools = llm.bind_tools( ^^^^^^^^^^^^^^^ File "/opt/render/project/src/.venv/lib/python3.11/site-packages/langchain_openai/chat_models/base.py", line 746, in bind_tools raise ValueError( ValueError: When specifying `tool_choice`, you must provide exactly one tool. Received 3 tools. ### Description I'm trying to let chatgpt decide between 3 custom functions by setting tool_choice equal to "auto" (as per the documentation), but it is falling into the error block for when the length of the functions list is != 1. This means the "auto" setting by which the ai chooses from more than 1 function is unreachable. Interestingly it works fine in my local environment in which I installed the libraries several weeks ago, but not in my cloud environment deployed today. I suspect it may have been a recent breaking change. Thanks guys, appreciate any help on this! ### System Info developed on mac OS, deployed to Render web service. Python 3.9.13

langchain_openai.chat_models.base.ChatOpenAI[source] bind_tools.tool_choice="auto" is not working.

https://api.github.com/repos/langchain-ai/langchain/issues/20103/comments

2024-04-06T11:11:15Z

2024-08-09T16:08:03Z

https://github.com/langchain-ai/langchain/issues/20103

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code model = ChatGroq( model_name="mixtral-8x7b-32768" ) class Joke(BaseModel): setup: str = Field(description="The setup of the joke") punchline: str = Field(description="The punchline to the joke") model_with_structure = model.with_structured_output(Joke, method="json_mode") f = model_with_structure.invoke( "Tell me a joke about cats, respond in JSON with `setup` and `punchline` keys" ) ### Error Message and Stack Trace (if applicable) python/3_11/venv/lib/python3.11/site-packages/langchain_core/_api/beta_decorator.py:87: LangChainBetaWarning: The function `with_structured_output` is in beta. It is actively being worked on, so the API may change. warn_beta( Traceback (most recent call last): File "/python/3_11/wp_app/src/aibro_langchain.py", line 25, in <module> model_with_structure = model.with_structured_output(Joke, method="json_mode") ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/python/3_11/venv/lib/python3.11/site-packages/langchain_core/_api/beta_decorator.py", line 110, in warning_emitting_wrapper return wrapped(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^ File "/python/3_11/venv/lib/python3.11/site-packages/langchain_core/language_models/base.py", line 204, in with_structured_output raise NotImplementedError() NotImplementedError ### Description Try to get structured output from groq ### System Info langchain==0.1.14 langchain-community==0.0.31 langchain-core==0.1.40 langchain-groq==0.0.1 langchain-openai==0.0.5 Mac python. 3.11

with_structured_output raise NotImplementedError() Version: 0.1.14

https://api.github.com/repos/langchain-ai/langchain/issues/20102/comments

2024-04-06T11:10:35Z

2024-06-25T03:30:28Z

https://github.com/langchain-ai/langchain/issues/20102

### Checklist - [X] I added a very descriptive title to this issue. - [X] I included a link to the documentation page I am referring to (if applicable). ### Issue with current documentation: https://python.langchain.com/docs/integrations/vectorstores/faiss/ Hello, I am currently working with langchain for document-related processing tasks, specifically utilizing the ```faiss.from_documents``` feature for indexing and similarity searches. I am interested in understanding what the default FAISS index type is when using faiss.from_documents without specifying any particular configuration. For instance, does it default to using PQIVF, LSH, or another type of index? After reading the source code, I find that I can only use inner product and L2 index. What should I do if I want to use IVFPQ? ### Idea or request for content: _No response_

DOC: < type of faiss index prefix>

https://api.github.com/repos/langchain-ai/langchain/issues/20097/comments

2024-04-06T02:38:41Z

2024-07-13T16:06:05Z

https://github.com/langchain-ai/langchain/issues/20097

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code import boto3 import json import os import langchain from langchain.llms.bedrock import Bedrock from langchain import hub from langchain.agents import AgentExecutor, create_structured_chat_agent from langchain_community.llms import Bedrock from langchain.tools import tool AWS_ACCESS_KEY = os.getenv('AWS_ACCESS_KEY_ID') AWS_SECRET_ACCESS_KEY = os.getenv('AWS_SECRET_ACCESS_KEY') AWS_REGION = os.getenv('AWS_REGION', 'us-east-1') bedrock = boto3.client( service_name='bedrock-runtime', aws_access_key_id=AWS_ACCESS_KEY, aws_secret_access_key=AWS_SECRET_ACCESS_KEY, region_name=AWS_REGION ) llm = Bedrock( credentials_profile_name="default", model_id="mistral.mistral-large-2402-v1:0") @tool def multiply(a: int, b: int): """Multiply two integers""" return a * b tools = [multiply] prompt = hub.pull("hwchase17/structured-chat-agent") agent = create_structured_chat_agent(llm, tools, prompt) agent_executor = AgentExecutor(agent=agent, tools=tools) result = agent_executor.invoke({"input": "what is 123 * 456"}) print(result) ### Error Message and Stack Trace (if applicable) File ".../env/lib/python3.11/site-packages/langchain_community/llms/bedrock.py", line 654, in _prepare_input_and_invoke_stream raise ValueError(f"Error raised by bedrock service: {e}") ValueError: Error raised by bedrock service: An error occurred (ValidationException) when calling the InvokeModelWithResponseStream operation: Malformed input request: #: extraneous key [stop_sequences] is not permitted, please reformat your input and try again. ### Description I'm using AWS Bedrock for an agent application. It throws an error due to a stop sequence parameter that isn't supported by the AWS api. The error can be mitigated by commenting out lines 611 - 619 in langchain_community.llms.bedrock # if stop: # if provider not in self.provider_stop_sequence_key_name_map: # raise ValueError( # f"Stop sequence key name for {provider} is not supported." # ) # # stop sequence from _generate() overrides # # stop sequences in the class attribute # _model_kwargs[self.provider_stop_sequence_key_name_map.get(provider)] = stop ### System Info langchain==0.1.14 langchain-community==0.0.31 langchain-core==0.1.40 langchain-openai==0.0.3 langchain-text-splitters==0.0.1 langchainhub==0.1.14 boto3==1.34.79 botocore==1.34.79

Stop Sequenced Not Supported by AWS Bedrock

https://api.github.com/repos/langchain-ai/langchain/issues/20095/comments

2024-04-06T00:57:13Z

2024-07-21T16:06:45Z

https://github.com/langchain-ai/langchain/issues/20095

### Checklist - [X] I added a very descriptive title to this issue. - [X] I included a link to the documentation page I am referring to (if applicable). ### Issue with current documentation: While reading and learning from the [LCEL - Getting Started](https://python.langchain.com/docs/expression_language/get_started/#rag-search-example) docs, I found one minor issue in the documentation. In RAG Search Example, ` (backtick) is missing around the keyword RunnablePassthrough - [GitHub Code](https://github.com/langchain-ai/langchain/blob/de496062b3e740aed2c7097424749d1145e4aaab/docs/docs/expression_language/get_started.ipynb#L443). Solution: RunnablePassthrough -> `RunnablePassthrough` ### Idea or request for content: _No response_

DOC: missing backtick around RunnablePassthrough on LCEL getting started

https://api.github.com/repos/langchain-ai/langchain/issues/20094/comments

2024-04-06T00:41:55Z

2024-04-11T12:39:24Z

https://github.com/langchain-ai/langchain/issues/20094

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ``` from langchain_community.document_loaders import DirectoryLoader, TextLoader loader = DirectoryLoader( path, glob="**/*.txt", use_multithreading=True, show_progress=True, sample_size=1000, loader_cls=TextLoader) docs: List[Document] = loader.load() ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description Expecting `List[Document]` but getting back `List[List[Document]]` which cannot be directly fed into `Chroma.from_documents` for example ### System Info langchain==0.1.14 langchain-community==0.0.31 langchain-core==0.1.38 langchain-openai==0.1.1 langchain-text-splitters==0.0.1 langchainhub==0.1.14 Windows Python 3.9.13

DirectoryLoader with use_multithreading=True loads each doc as a list of docs

https://api.github.com/repos/langchain-ai/langchain/issues/20093/comments

2024-04-06T00:32:34Z

2024-07-13T16:06:00Z

https://github.com/langchain-ai/langchain/issues/20093

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code libs/community/langchain_community/document_loaders/confluence.py @@ -359,6 +359,7 @@ def _lazy_load(self, **kwargs: Any) -> Iterator[Document]: content_format, ocr_languages, keep_markdown_format, **keep_newlines=keep_newlines** ) ### Error Message and Stack Trace (if applicable) _No response_ ### Description I use the confluence loader of langchain to download the pages content of a specific page of my confluence instance. While textspllitting/chunking the pages I've noticed that in none-markdown format the newlines were missing. During the debugging I saw that that the required forward-pass of the `keep_newlines` parameter was not passed down to all call of the `process_pages` function inside of `libs/community/langchain_community/document_loaders/confluence.py` ### System Info langchain=0.1.14 windows 11 python 3.10

Confluence loader "keep_newlines" not always passed to "process_pages"

https://api.github.com/repos/langchain-ai/langchain/issues/20086/comments

2024-04-05T20:31:46Z

2024-07-12T16:04:49Z

https://github.com/langchain-ai/langchain/issues/20086

### Privileged issue - [X] I am a LangChain maintainer, or was asked directly by a LangChain maintainer to create an issue here. ### Issue Content There's some discrepancies in the init args different models use to set the same params. It'd be a much nicer UX if common params could be set with a common set of init args Suggesting that if a param in this list is present in a model integration, the below name should either be the actual attribute name or an init arg that can be used to set the attribute: ```python model: str # model name api_key: str # api key temperature: float # temperature sampling timeout: ... # request timeout max_tokens: int # max tokens stop: ... # stop sequences max_retries: int # max num retries base_url: str # base URL to send requests to ``` Importantly we should also use the above init args in the docs

Standardized model init arg names

https://api.github.com/repos/langchain-ai/langchain/issues/20085/comments

2024-04-05T20:30:47Z

2024-06-10T12:49:39Z

https://github.com/langchain-ai/langchain/issues/20085

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python import os from langchain_community.llms import Ollama from langchain.agents import AgentType, initialize_agent from langchain_community.agent_toolkits.jira.toolkit import JiraToolkit from langchain_community.utilities.jira import JiraAPIWrapper from langchain_core.prompts import ChatPromptTemplate os.environ["JIRA_API_TOKEN"] = "token" os.environ["JIRA_USERNAME"] = "user" os.environ["JIRA_INSTANCE_URL"] = "https://jira.atlassian.net" llm = Ollama(model="llama2:13b") jira = JiraAPIWrapper() toolkit = JiraToolkit.from_jira_api_wrapper(jira) agent = initialize_agent( toolkit.get_tools(), llm, agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION, verbose=True ) # chain = prompt | chat resp = agent.invoke( "make a new issue in project id 'MUG' to remind me to make more fried rice" ) ``` ### Error Message and Stack Trace (if applicable) ```bash LangChainDeprecationWarning: The function `initialize_agent` was deprecated in LangChain 0.1.0 and will be removed in 0.2.0. Use Use new agent constructor methods like create_react_agent, create_json_agent, create_structured_chat_agent, etc. instead. warn_deprecated( > Entering new AgentExecutor chain... Question: make a new issue in project id 'MUG' to remind me to make more fried rice Thought: I should create a new issue in the MUG project to remind myself to make more fried rice. Action: Create Issue Action Input: { "summary": "Reminder to make more fried rice", "description": "I want to make more fried rice, this is a reminder.", "issuetype": {"name": "Task"}, "priority": {"name": "Low"} }Creating issue "Reminder to make more fried rice" Traceback (most recent call last): File "/Users/samuelbirocchi/.pyenv/versions/3.10.13/lib/python3.10/runpy.py", line 196, in _run_module_as_main return _run_code(code, main_globals, None, File "/Users/samuelbirocchi/.pyenv/versions/3.10.13/lib/python3.10/runpy.py", line 86, in _run_code exec(code, run_globals) File "/Users/samuelbirocchi/.vscode/extensions/ms-python.debugpy-2024.4.0-darwin-arm64/bundled/libs/debugpy/adapter/../../debugpy/launcher/../../debugpy/__main__.py", line 39, in <module> cli.main() File "/Users/samuelbirocchi/.vscode/extensions/ms-python.debugpy-2024.4.0-darwin-arm64/bundled/libs/debugpy/adapter/../../debugpy/launcher/../../debugpy/../debugpy/server/cli.py", line 430, in main run() File "/Users/samuelbirocchi/.vscode/extensions/ms-python.debugpy-2024.4.0-darwin-arm64/bundled/libs/debugpy/adapter/../../debugpy/launcher/../../debugpy/../debugpy/server/cli.py", line 284, in run_file runpy.run_path(target, run_name="__main__") File "/Users/samuelbirocchi/.vscode/extensions/ms-python.debugpy-2024.4.0-darwin-arm64/bundled/libs/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 321, in run_path return _run_module_code(code, init_globals, run_name, File "/Users/samuelbirocchi/.vscode/extensions/ms-python.debugpy-2024.4.0-darwin-arm64/bundled/libs/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 135, in _run_module_code _run_code(code, mod_globals, init_globals, File "/Users/samuelbirocchi/.vscode/extensions/ms-python.debugpy-2024.4.0-darwin-arm64/bundled/libs/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 124, in _run_code exec(code, run_globals) File "./main.py", line 45, in <module> resp = agent.invoke( File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain/chains/base.py", line 163, in invoke raise e File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain/chains/base.py", line 153, in invoke self._call(inputs, run_manager=run_manager) File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain/agents/agent.py", line 1432, in _call next_step_output = self._take_next_step( File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain/agents/agent.py", line 1138, in _take_next_step [ File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain/agents/agent.py", line 1138, in <listcomp> [ File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain/agents/agent.py", line 1223, in _iter_next_step yield self._perform_agent_action( File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain/agents/agent.py", line 1245, in _perform_agent_action observation = tool.run( File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain_core/tools.py", line 422, in run raise e File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain_core/tools.py", line 381, in run self._run(*tool_args, run_manager=run_manager, **tool_kwargs) File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain_community/tools/jira/tool.py", line 44, in _run return self.api_wrapper.run(self.mode, instructions) File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain_community/utilities/jira.py", line 168, in run return self.issue_create(query) File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/langchain_community/utilities/jira.py", line 139, in issue_create return self.jira.issue_create(fields=dict(params)) File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/atlassian/jira.py", line 1537, in issue_create return self.post(url, data={"fields": fields}) File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/atlassian/rest_client.py", line 388, in post response = self.request( File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/atlassian/rest_client.py", line 312, in request self.raise_for_status(response) File "/Users/samuelbirocchi/langchain/.venv/lib/python3.10/site-packages/atlassian/rest_client.py", line 549, in raise_for_status raise HTTPError(error_msg, response=response) requests.exceptions.HTTPError ``` ### Description I'm trying to create a Jira ticket following the documentation ### System Info langchain==0.1.14 langchain-community==0.0.31 langchain-core==0.1.40 langchain-text-splitters==0.0.1 Python 3.10.13

Jira toolkit prompt for create ticket does not include project ID

https://api.github.com/repos/langchain-ai/langchain/issues/20084/comments

2024-04-05T20:11:01Z

2024-06-05T04:14:30Z

https://github.com/langchain-ai/langchain/issues/20084

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code In the example below, if the `keep_separator` parameter is set to `True`, the separator's position will appear at the beginning of each chunk, whereas we would prefer it to appear at the end. ```python from langchain_text_splitters import CharacterTextSplitter content = "Hello world! Nice to meet you! Nice to meet you too!" text_splitter = CharacterTextSplitter(separator="!", chunk_size=25, chunk_overlap=0, keep_separator=True) chunks = text_splitter.create_documents([content]) for chunk in chunks: print(chunk.page_content) ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description In this case, my raw input is `Hello world! Nice to meet you! Nice to meet you too!`, and my expected output is: ``` # Expected Hello world! Nice to meet you! Nice to meet you too! ``` However, the real output is: ``` # Real Hello world ! Nice to meet you ! Nice to meet you too! ``` --- At first, I thought it was a bug, so I fixed the bug and was ready to `create pull request`. ```python # venv/Lib/site-packages/langchain_text_splitters/character.py def _split_text_with_regex(...): ... # splits = [_splits[i] + _splits[i + 1] for i in range(1, len(_splits), 2)] # if len(_splits) % 2 == 0: # splits += _splits[-1:] # splits = [_splits[0]] + splits # --------replace with below-------- if len(_splits) % 2 != 0: _splits.append("") splits = [_splits[i] + _splits[i + 1] for i in range(0, len(_splits), 2)] ... ``` However, I found that it couldn't pass the test. Upon reviewing the test code, I discovered that this was not a mistake by the author, which left me confused. @baskaryan @hwchase17 ### System Info (venv) ~\Desktop\workspace\python\langchain-imooc git:[main] python -m langchain_core.sys_info System Information ------------------ > OS: Windows > OS Version: 10.0.22631 > Python Version: 3.12.2 (tags/v3.12.2:6abddd9, Feb 6 2024, 21:26:36) [MSC v.1937 64 bit (AMD64)] Package Information ------------------- > langchain_core: 0.1.37 > langchain: 0.1.14 > langchain_community: 0.0.30 > langsmith: 0.1.38 > langchain_openai: 0.1.1 > langchain_text_splitters: 0.0.1 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve

Confusion about TextSplitter (param keep_separator)

https://api.github.com/repos/langchain-ai/langchain/issues/20066/comments

2024-04-05T13:50:38Z

2024-07-12T16:04:43Z

https://github.com/langchain-ai/langchain/issues/20066

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code SemanticSimilarityExampleSelector(vectorstore=vectorstore, k=2) is throwing an error : > File [c:\Users\l1017\.conda\envs\llm\Lib\site-packages\pydantic\main.py:341](file:///C:/Users/l1017/.conda/envs/llm/Lib/site-packages/pydantic/main.py:341), in pydantic.main.BaseModel.__init__() ValidationError: 1 validation error for SemanticSimilarityExampleSelector vectorstore instance of VectorStore expected (type=type_error.arbitrary_type; expected_arbitrary_type=VectorStore) My code below: <img width="809" alt="微信图片_20240405171258" src="https://github.com/langchain-ai/langchain/assets/37232258/5b9afdb6-9058-4f1e-8a0b-ada02634a290"> ### Error Message and Stack Trace (if applicable) File [c:\Users\l1017\.conda\envs\llm\Lib\site-packages\pydantic\main.py:341](file:///C:/Users/l1017/.conda/envs/llm/Lib/site-packages/pydantic/main.py:341), in pydantic.main.BaseModel.__init__() ValidationError: 1 validation error for SemanticSimilarityExampleSelector vectorstore instance of VectorStore expected (type=type_error.arbitrary_type; expected_arbitrary_type=VectorStore) ### Description I upgrade the langchain to the lastest version,but not work ### System Info langchain 0.1.40 platform - windows 11 python version - 3.12.2

It returns "instance of VectorStore expected" error, when I creating SemanticSimilarityExampleSelector instance

https://api.github.com/repos/langchain-ai/langchain/issues/20062/comments

2024-04-05T12:05:48Z

2024-07-12T16:04:38Z

https://github.com/langchain-ai/langchain/issues/20062

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code This is the code: ```from langchain_core.prompts import ChatPromptTemplate from langchain_core.runnables import ConfigurableField from langchain_anthropic import ChatAnthropic model = ChatAnthropic( model="claude-3-sonnet-20240229", max_tokens=1024, temperature=0, anthropic_api_key="<api_key>" ).configurable_fields( max_tokens=ConfigurableField( id="max_tokens", name="Number of LLM tokens", description="Number of LLM tokens" ) ) prompt = ChatPromptTemplate.from_template("Write an essay on Tiger.") chain = prompt | model # print(chain.invoke({})) # chain.invoke({"x": "Indian Constitution"}) print(chain.invoke({}, config={"configurable": {"max_tokens": 100}}))``` ### Error Message and Stack Trace (if applicable) Warnings: ```/home/mayank/miniconda3/envs/jupyter/lib/python3.9/site-packages/langchain_core/utils/utils.py:159: UserWarning: WARNING! _client is not default parameter. _client was transferred to model_kwargs. Please confirm that _client is what you intended. warnings.warn( /home/mayank/miniconda3/envs/jupyter/lib/python3.9/site-packages/langchain_core/utils/utils.py:159: UserWarning: WARNING! _async_client is not default parameter. _async_client was transferred to model_kwargs. Please confirm that _async_client is what you intended. warnings.warn( {'model': 'claude-3-sonnet-20240229', 'max_tokens': 100, 'messages': [{'role': 'user', 'content': 'Write an essay on Tiger.'}], 'temperature': 0.0, '_client': <anthropic.Anthropic object at 0x7c1b3cc80c70>, '_async_client': <anthropic.AsyncAnthropic object at 0x7c1b360b1e50>}``` Traceback (most recent call last): ```File "/home/mayank/open-source/test-notebooks/./test.py", line 21, in <module> print(chain.invoke({}, config={"configurable": {"max_tokens": 100}})) File "/home/mayank/miniconda3/envs/jupyter/lib/python3.9/site-packages/langchain_core/runnables/base.py", line 2499, in invoke input = step.invoke( File "/home/mayank/miniconda3/envs/jupyter/lib/python3.9/site-packages/langchain_core/runnables/configurable.py", line 95, in invoke return runnable.invoke(input, config, **kwargs) File "/home/mayank/miniconda3/envs/jupyter/lib/python3.9/site-packages/langchain_core/language_models/chat_models.py", line 154, in invoke self.generate_prompt( File "/home/mayank/miniconda3/envs/jupyter/lib/python3.9/site-packages/langchain_core/language_models/chat_models.py", line 554, in generate_prompt return self.generate(prompt_messages, stop=stop, callbacks=callbacks, **kwargs) File "/home/mayank/miniconda3/envs/jupyter/lib/python3.9/site-packages/langchain_core/language_models/chat_models.py", line 415, in generate raise e File "/home/mayank/miniconda3/envs/jupyter/lib/python3.9/site-packages/langchain_core/language_models/chat_models.py", line 405, in generate self._generate_with_cache( File "/home/mayank/miniconda3/envs/jupyter/lib/python3.9/site-packages/langchain_core/language_models/chat_models.py", line 624, in _generate_with_cache result = self._generate( File "/home/mayank/miniconda3/envs/jupyter/lib/python3.9/site-packages/langchain_anthropic/chat_models.py", line 292, in _generate data = self._client.messages.create(**params) File "/home/mayank/miniconda3/envs/jupyter/lib/python3.9/site-packages/anthropic/_utils/_utils.py", line 275, in wrapper return func(*args, **kwargs) TypeError: create() got an unexpected keyword argument '_client'``` ### Description It was expected to change `max_tokens` and the generate the response ### System Info ```System Information ------------------ > OS: Linux > OS Version: #26~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Mar 12 10:22:43 UTC 2 > Python Version: 3.9.19 (main, Mar 21 2024, 17:11:28) [GCC 11.2.0] Package Information ------------------- > langchain_core: 0.1.40 > langchain: 0.1.13 > langchain_community: 0.0.29 > langsmith: 0.1.38 > langchain_anthropic: 0.1.4 > langchain_openai: 0.1.1 > langchain_text_splitters: 0.0.1 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve```

ChatAnthropic gives error while invoking it after instantiating with `configurable_fields`

https://api.github.com/repos/langchain-ai/langchain/issues/20061/comments

2024-04-05T11:40:10Z

2024-07-12T16:04:34Z

https://github.com/langchain-ai/langchain/issues/20061

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code I am trying to use AWS Bedrock models such as Llama / Mistral with Langchain Libraries such as SQLDatabaseToolkit. ``` model = Bedrock(credentials_profile_name="my-profile", model_id="meta.llama2-70b-chat-v1", model_kwargs={"temperature": 0.5}, streaming=True, callbacks=[StreamingStdOutCallbackHandler()]) db = SQLDatabase.from_uri('database url') toolkit = SQLDatabaseToolkit(llm=model,db=db) agent_executor = create_sql_agent(llm=model, toolkit=toolkit,verbose=True, handle_parsing_errors=True) # Query Module result = agent_executor.invoke(prompt) ``` ### Error Message and Stack Trace (if applicable) This errors out with the following. ``` File "/opt/homebrew/lib/python3.11/site-packages/langchain_community/llms/bedrock.py", line 833, in _call for chunk in self._stream( File "/opt/homebrew/lib/python3.11/site-packages/langchain_community/llms/bedrock.py", line 613, in _prepare_input_and_invoke_stream raise ValueError( ValueError: Stop sequence key name for meta is not supported. ``` ### Description I have tried the same code with `OpenAI` / `Ollama Mistral/Lamma` as well as google GenAI models and they don't seem to show this error. This seems like something with the way the bedrock library works in Langchain or the bedrock service. Is there a workaround I can use to get this to work. ### System Info System Information ------------------ > OS: Darwin > OS Version: Darwin Kernel Version 23.3.0: Wed Dec 20 21:30:44 PST 2023 > Python Version: 3.11.7 (main, Dec 4 2023, 18:10:11) [Clang 15.0.0 (clang-1500.1.0.2.5)] Package Information ------------------- > langchain_core: 0.1.40 > langchain: 0.1.14 > langchain_community: 0.0.31 > langsmith: 0.1.40 > langchain_experimental: 0.0.56 > langchain_openai: 0.0.5 > langchain_text_splitters: 0.0.1 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve

Error : Stop sequence key name for {meta or mistral or any other mode} is not supported with AWS Bedrock

https://api.github.com/repos/langchain-ai/langchain/issues/20053/comments

2024-04-05T06:43:08Z

2024-07-21T13:56:17Z

https://github.com/langchain-ai/langchain/issues/20053

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python import pydantic.v1 from langchain_core.tools import tool class MySchema(pydantic.v1.BaseModel): sum: bool # whoops @tool def add(a: int, b: int) -> MySchema: """ Add two numbers """ return MySchema(sum=a + b) add.handle_validation_error = '-1' print(add({'a': 3, 'b': 5})) # prints -1, but should raise an exception ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description The [`try: ... except ValidationError as e:` block in `BaseTool.run`](https://github.com/langchain-ai/langchain/blob/master/libs/core/langchain_core/tools.py#L377-L385) wraps the call to `self._run`. I assume it's supposed to capture errors raised in `_parse_input` only, but currently captures validation errors raised inside `self._run` as well. This might hide bugs in tool implementations that use `pydantic` when `handle_validation_error` is not `None`. ### System Info langchain==0.1.11 langchain-community==0.0.27 langchain-core==0.1.30 langchain-text-splitters==0.0.1

`pydantic.ValidationError` raised inside the `_run` method of a tool should not be captured by `handle_validation_error`

https://api.github.com/repos/langchain-ai/langchain/issues/20045/comments

2024-04-05T01:53:08Z

2024-07-12T16:04:28Z

https://github.com/langchain-ai/langchain/issues/20045

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python from langchain_core.callbacks import BaseCallbackHandler from langchain_core.tools import tool @tool def add(a: int, b: int) -> int: """ Add two numbers """ return a + b class MyCallback(BaseCallbackHandler): def on_tool_end(self, output: str, **kwargs): print(output) add({'a': 2, 'b': 2}, callbacks=[MyCallback()]) # prints 4 add.handle_validation_error = '-1' add({'a': '3'}, callbacks=[MyCallback()]) # does not print anything, supposed to print -1 ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description Unlike other execution paths of `BaseTool.run` that return an observation, [the path that handles validation errors](https://github.com/langchain-ai/langchain/blob/master/libs/core/langchain_core/tools.py#L386-L399) does not call `run_manager.on_tool_end` callback method. It prevents users from accessing validation error observations in custom callbacks. ### System Info langchain==0.1.11 langchain-community==0.0.27 langchain-core==0.1.30 langchain-text-splitters==0.0.1

`run_manager.on_tool_end` is not called with the output of `handle_validation_error` in `BaseTool.run`

https://api.github.com/repos/langchain-ai/langchain/issues/20043/comments

2024-04-05T01:44:26Z

2024-07-12T16:04:23Z

https://github.com/langchain-ai/langchain/issues/20043

### Checklist - [X] I added a very descriptive title to this issue. - [X] I included a link to the documentation page I am referring to (if applicable). ### Issue with current documentation: how to find what default prompt used for LLMGraphTransformer from langchain_experimental.graph_transformers import LLMGraphTransformer in class description it is not described what default prompt is class LLMGraphTransformer: """Transform documents into graph-based documents using a LLM. It allows specifying constraints on the types of nodes and relationships to include in the output graph. The class doesn't support neither extract and node or relationship properties Args: llm (BaseLanguageModel): An instance of a language model supporting structured output. allowed_nodes (List[str], optional): Specifies which node types are allowed in the graph. Defaults to an empty list, allowing all node types. allowed_relationships (List[str], optional): Specifies which relationship types are allowed in the graph. Defaults to an empty list, allowing all relationship types. prompt (Optional[ChatPromptTemplate], optional): The prompt to pass to the LLM with additional instructions. strict_mode (bool, optional): Determines whether the transformer should apply filtering to strictly adhere to `allowed_nodes` and `allowed_relationships`. Defaults to True. Example: .. code-block:: python from langchain_experimental.graph_transformers import LLMGraphTransformer from langchain_core.documents import Document from langchain_openai import ChatOpenAI llm=ChatOpenAI(temperature=0) transformer = LLMGraphTransformer( llm=llm, allowed_nodes=["Person", "Organization"]) doc = Document(page_content="Elon Musk is suing OpenAI") graph_documents = transformer.convert_to_graph_documents([doc]) """ ### Idea or request for content: add to class description what is default prompt is

DOC: how to find what default prompt used for LLMGraphTransformer

https://api.github.com/repos/langchain-ai/langchain/issues/20020/comments

2024-04-04T20:29:55Z

2024-07-12T16:04:19Z

https://github.com/langchain-ai/langchain/issues/20020

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code I got this error when calling this method agent_executor.invoke({"input": "hi!"}) I used Azure LLM like below : # Import Azure OpenAI from langchain_openai import AzureOpenAI # Create an instance of Azure OpenAI With : llm = AzureOpenAI( deployment_name="gpt-35-turbo-1106", ) ### Error Message and Stack Trace (if applicable) _No response_ ### Description We now enter the agent loop (until it returns something). 1390 while self._should_continue(iterations, time_elapsed): -> 1391 next_step_output = self._take_next_step( 1392 name_to_tool_map, 1393 color_mapping, 1394 inputs, 1395 intermediate_steps, 1396 run_manager=run_manager, 1397 ) 1398 if isinstance(next_step_output, AgentFinish): 1399 return self._return( 1400 next_step_output, intermediate_steps, run_manager=run_manager 1401 ) in AgentExecutor._take_next_step(self, name_to_tool_map, color_mapping, inputs, intermediate_steps, run_manager) ... 273 msg = f"Missing required argument: {quote(missing[0])}" 274 raise TypeError(msg) --> 275 return func(*args, **kwargs) TypeError: Completions.create() got an unexpected keyword argument 'functions' ### System Info Python version IS 3.11 Langchain IS the last one

Testing Quickstart agents with Azure open ai

https://api.github.com/repos/langchain-ai/langchain/issues/20013/comments

2024-04-04T17:30:25Z

2024-07-11T16:07:19Z

https://github.com/langchain-ai/langchain/issues/20013

### Checklist - [X] I added a very descriptive title to this issue. - [X] I included a link to the documentation page I am referring to (if applicable). ### Issue with current documentation: `langchain_community.llms.huggingface_hub.HuggingFaceHub` was deprecated, and `langchain_community.chat_models.huggingface.ChatHuggingFace` only Works with HuggingFaceTextGenInference, HuggingFaceEndpoint, and HuggingFaceHub So how to create a ChatHuggingFace instance from local huggingface Model Now?? source :https://api.python.langchain.com/en/latest/chat_models/langchain_community.chat_models.huggingface.ChatHuggingFace.html#langchain_community.chat_models.huggingface.ChatHuggingFace https://api.python.langchain.com/en/latest/llms/langchain_community.llms.huggingface_hub.HuggingFaceHub.html#langchain_community.llms.huggingface_hub.HuggingFaceHub ### Idea or request for content: document says HuggingFaceHub was deprecated, i should use the HuggingFaceEndPoint, but that is two difference integrate method.

DOC: How to create a ChatModel from Huggingface using local llm

https://api.github.com/repos/langchain-ai/langchain/issues/19997/comments

2024-04-04T11:22:15Z

2024-05-22T03:32:01Z

https://github.com/langchain-ai/langchain/issues/19997

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```py from langchain_openai import AzureChatOpenAI llm = AzureChatOpenAI( azure_endpoint=openai_api_base, # type: ignore openai_api_version=openai_api_version, # type: ignore azure_deployment=deployment_name, openai_api_key=openai_api_key, # type: ignore openai_api_type=openai_api_type, # type: ignore temperature=openai_llm_temp, callbacks=[handler], model_name="gpt-4-32k", openai_proxy=openai_proxy ) ``` ### Error Message and Stack Trace (if applicable) No exception or full stack trace, simply requests being timed out because they don't go thru our corporate proxy. ### Description I'm trying to tell AzureChatOpenAI to use our corporate proxy, however under langchain-openai it doesn't seem to take it in account. I've had to downgrade to use AzureChatOpenAI in langchain and downgrade the OpenAI package to respectively: langchain - 0.1.14 openai - 0.28.1 ### System Info ``` System Information ------------------ > OS: Darwin > OS Version: Darwin Kernel Version 23.3.0: Wed Dec 20 21:31:00 PST 2023; root:xnu-10002.81.5~7/RELEASE_ARM64_T6020 > Python Version: 3.11.7 (main, Dec 4 2023, 18:10:11) [Clang 15.0.0 (clang-1500.1.0.2.5)] Package Information ------------------- > langchain_core: 0.1.39 > langchain: 0.1.14 > langchain_community: 0.0.31 > langsmith: 0.1.39 > langchain_openai: 0.0.8 > langchain_text_splitters: 0.0.1 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve ```

Azure ChatOpenAI won't use proxy provided either via env var or value

https://api.github.com/repos/langchain-ai/langchain/issues/19994/comments

2024-04-04T08:27:43Z

2024-05-30T10:04:15Z

https://github.com/langchain-ai/langchain/issues/19994

### Checklist - [X] I added a very descriptive title to this issue. - [X] I included a link to the documentation page I am referring to (if applicable). ### Issue with current documentation: [This page](https://python.langchain.com/docs/modules/model_io/output_parsers/) provides a summary table of langchain output parsers. However, when one compares this with [the API page](https://api.python.langchain.com/en/latest/langchain_api_reference.html#module-langchain.output_parsers), [the API page for langchain_core](https://api.python.langchain.com/en/latest/core_api_reference.html#module-langchain_core.output_parsers) and [the output page for langchain_community](https://api.python.langchain.com/en/latest/community_api_reference.html#module-langchain_community.output_parsers) many, many output parsers are not documented. It would also be a big help if there were links from the class descriptions on this page to the class definitions in the API docs (in general, adding cross-links to the API docs would be a huge improvement to the langchain docs. ### Idea or request for content: 1. Cover a full set of the output parsers in the documentation page dedicated to them. 2. Add more cross-links.

DOC: Documentation of output parsers is incomplete

https://api.github.com/repos/langchain-ai/langchain/issues/19969/comments

2024-04-03T19:25:39Z

2024-07-10T16:07:05Z

https://github.com/langchain-ai/langchain/issues/19969

@eyurtsev The current workaround doesn't work for create_sql_agent which is a toolkit type. Can you please provide a fix for this as well? I agreed with most users - token counting is one of the fundamental and crucial features by LangChain but this doesn't seem to be treated as a high priority. @hwchase17 Thanks. Please help the community. _Originally posted by @axtsoi in https://github.com/langchain-ai/langchain/issues/18761#issuecomment-2033491551_

Add option to turn off streaming in sql agent

https://api.github.com/repos/langchain-ai/langchain/issues/19961/comments

2024-04-03T16:11:53Z

2024-04-09T16:47:34Z

https://github.com/langchain-ai/langchain/issues/19961

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code I have this code in my Dockerfile: RUN pip install --no-cache-dir fastapi uvicorn redis requests tiktoken openai And here are my dependecies : fastapi==0.110.1 langchain==0.1.14 openai==1.16.1 pydantic==1.10.9 python-dotenv==1.0.1 redis==5.0.3 Requests==2.31.0 tiktoken==0.6.0 ### Error Message and Stack Trace (if applicable) File "/usr/local/lib/python3.12/site-packages/langchain_community/embeddings/openai.py", line 452, in _get_len_safe_embeddings raise ImportError(ImportError: Could not import tiktoken python package. This is needed in order to for OpenAIEmbeddings. Please install it with `pip install tiktoken`. ### Description I am trying to deploy a RAG application using langchain, everything works fine locally and I installed tiktoken. But when I try to deploy I get an error message saying it cannot import tiktoken. Do you have any clue on how I can solve this issue ? ### System Info Python 3.11.4 langchain==0.1.14 langchain-community==0.0.31 langchain-core==0.1.39 langchain-experimental==0.0.37 langchain-openai==0.1.1 langchain-text-splitters==0.0.1

unable to import tiktoken

https://api.github.com/repos/langchain-ai/langchain/issues/19958/comments

2024-04-03T14:16:14Z

2024-04-03T14:47:31Z

https://github.com/langchain-ai/langchain/issues/19958

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python from langchain.chains import LLMChain from langchain.llms import OpenAI from langchain.prompts import PromptTemplate llm = OpenAI() prompt = PromptTemplate( input_variables=["product"], template="What is a good name for a company that makes {product}?", ) chain = LLMChain(llm=llm, prompt=prompt) chain.save("chain.json") from langchain.chains.load_chain chain = load_chain("chain.json") ``` One liner to reproduce this error: ``` docker run --rm python:3.9 bash -c 'pip install langchain && python -c "from langchain.chains import load_chain; load_chain(\"/path/to/chain.json\")"' ``` ### Error Message and Stack Trace (if applicable) ``` Traceback (most recent call last): File "//a.py", line 16, in <module> chain = load_chain("chain.json") File "/usr/local/lib/python3.9/site-packages/langchain/chains/loading.py", line 625, in load_chain if hub_result := try_load_from_hub( File "/usr/local/lib/python3.9/site-packages/langchain_core/utils/loading.py", line 10, in try_load_from_hub raise RuntimeError( RuntimeError: Loading from the deprecated github-based Hub is no longer supported. Please use the new LangChain Hub at https://smith.langchain.com/hub instead. ``` ### Description `langchain.chains.load_chain` fails with `RuntimeError: Loading from the deprecated github-based Hub is no longer supported. Please use the new LangChain Hub at https://smith.langchain.com/hub instead.`. ### System Info ``` # pip list | grep langchain langchain 0.1.14 langchain-community 0.0.31 langchain-core 0.1.39 langchain-text-splitters 0.0.1 ```

`langchain.chains.load_chain` fails with `RuntimeError: Loading from the deprecated github-based Hub ...` in langchain 0.1.14 and langchain-core 0.1.39

https://api.github.com/repos/langchain-ai/langchain/issues/19947/comments

2024-04-03T07:06:26Z

2024-04-03T17:00:27Z

https://github.com/langchain-ai/langchain/issues/19947

### Checklist - [x] I added a very descriptive title to this issue. - [x] I included a link to the documentation page I am referring to (if applicable). ### Issue with current documentation:  Currently, calling batch will output the results all at once. How can I obtain streaming output while calling batch? For example, my input has 10 lines, and each time I get the first few words in the 10 lines until each line is completed. ### Idea or request for content: _No response_

DOC: How to use stream output while calling batch

https://api.github.com/repos/langchain-ai/langchain/issues/19944/comments

2024-04-03T05:42:34Z

2024-04-03T14:43:51Z

https://github.com/langchain-ai/langchain/issues/19944

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python import os from langchain_community.document_loaders.sharepoint import SharePointLoader os.environ['O365_CLIENT_ID'] = "<client_id>" os.environ['O365_CLIENT_SECRET'] = "<client_secret>" # option 1: this returns empty list loader = SharePointLoader(document_library_id="<document_library_id>", auth_with_token=True) loader.load() # option 2: this returns correctly, but only documents with listed ids loader = SharePointLoader(document_library_id="<document_library_id>", object_ids=["object_id_1","object_id_2"], auth_with_token=True) loader.load() ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description ## I'm trying to use [SharePoint document loader](https://github.com/langchain-ai/langchain/blob/master/libs/community/langchain_community/document_loaders/sharepoint.py). - When I don't pass in any `folder_path` I would expect the loader to load all supported documents from the SharePoint. However I get empty list `[]`. There are supported documents in the root directory. - Any `folder_path` I try to pass in is rejected with `ValueError: There isn't a folder with path <folder_path>`. It may have something to do with the root directory being called `Shared Documents`. The space in the folder name might be causing trouble. _*more details in a comment_ - When I pass in `object_id`s, I get the documents returned correctly. ### System Info ``` langchain==0.1.14 langchain-community==0.0.31 langchain-core==0.1.38 langchain-openai==0.0.8 langchain-text-splitters==0.0.1 Windows / Python 3.11.5 ```

SharePoint loader returns empty list when `object_ids` are not passed in

https://api.github.com/repos/langchain-ai/langchain/issues/19938/comments

2024-04-02T22:31:05Z

2024-07-09T16:07:54Z

https://github.com/langchain-ai/langchain/issues/19938

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code loader = GithubFileLoader( repo="hotgluexyz/tap-netsuite", access_token=gh_token, github_api_url="https://api.github.com", file_filter=lambda file_path: file_path.endswith( ".py" ), # retrieves all python files ) documents = loader.load() ### Error Message and Stack Trace (if applicable) --------------------------------------------------------------------------- HTTPError Traceback (most recent call last) [<ipython-input-7-5f7842728e5f>](https://localhost:8080/#) in <cell line: 1>() ----> 1 documents = loader.load() 3 frames [/usr/local/lib/python3.10/dist-packages/requests/models.py](https://localhost:8080/#) in raise_for_status(self) 1019 1020 if http_error_msg: -> 1021 raise HTTPError(http_error_msg, response=self) 1022 1023 def close(self): HTTPError: 404 Client Error: Not Found for url: https://api.github.com/repos/hotgluexyz/tap-netsuite/git/trees/main?recursive=1 ### Description I am trying to use langchain to load a github repository according to the documentation ### System Info langchain==0.1.14 langchain-community==0.0.31 langchain-core==0.1.38 langchain-openai==0.1.1 langchain-text-splitters==0.0.1 langcodes==3.3.0 langsmith==0.1.38

GithubFileLoader only imports repositories with main branch

https://api.github.com/repos/langchain-ai/langchain/issues/19934/comments

2024-04-02T21:49:56Z

2024-04-03T15:27:22Z

https://github.com/langchain-ai/langchain/issues/19934

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code Hugging Face pipeline now has [support for chat templates](https://huggingface.co/docs/transformers/main/en/chat_templating#is-there-an-automated-pipeline-for-chat). This calls the ``apply_chat_template()`` of the tokenizer. This is a super useful feature which formats the input correctly according to the model. To apply the template one needs to pass a messages list to the pipeline as input (and not a prompt text). Langchain's ``HuggingFacePipeline`` class is written in a way that only prompt text is passed to the pipeline. We can see this in the ``‎HuggingFacePipeline._generate`` method. As a result the prompt is constructed using Langchain's default template which is not the same as what the model works best with. Let's build an example. ```python import torch from langchain_core.prompts import ChatPromptTemplate from langchain_core.output_parsers import StrOutputParser from langchain_community.llms.huggingface_pipeline import HuggingFacePipeline def test_agnostic_prompt(llm): prompt = ChatPromptTemplate.from_messages( [ ("human", "When was Abraham Lincoln born?"), ("ai", "Abraham Lincoln was born on February 12, 1809."), ("human", "How old was he when he died?"), ("ai", "Abraham Lincoln died on April 15, 1865, at the age of 56."), ("human", "{question}"), ] ) output_parser = StrOutputParser() chain = prompt | llm | output_parser reply = chain.invoke({"question": "Where did he die?"}) print(reply) hf_llm = HuggingFacePipeline.from_model_id( model_id="TinyLlama/TinyLlama-1.1B-Chat-v1.0", task="text-generation", pipeline_kwargs={"max_new_tokens": 128}) test_agnostic_prompt(hf_llm) ``` This sends the following prompt. ``` Human: When was Abraham Lincoln born? AI: Abraham Lincoln was born on February 12, 1809. Human: How old was he when he died? AI: Abraham Lincoln died on April 15, 1865, at the age of 56. Human: Where did he die? ``` The correct prompt, if chat template was applied, would be: ``` <|user|> When was Abraham Lincoln born?</s> <|assistant|> Abraham Lincoln was born on February 12, 1809.</s> <|user|> How old was he when he died?</s> <|assistant|> Abraham Lincoln died on April 15, 1865, at the age of 56.</s> <|user|> Where did he die?</s> <|assistant|> ``` ### Error Message and Stack Trace (if applicable) _No response_ ### Description The ``HuggingFacePipeline`` class should what is necessary to convert the ``ChatPromptTemplate`` to a messages list and then pass it to the pipeline. This will cause the pipeline to use ``apply_chat_template()`` of the tokenizer to correctly format the prompt. ### System Info System Information ------------------ > OS: Linux > OS Version: #1 SMP Sat Feb 24 09:50:35 UTC 2024 > Python Version: 3.10.13 | packaged by conda-forge | (main, Dec 23 2023, 15:36:39) [GCC 12.3.0] Package Information ------------------- > langchain_core: 0.1.38 > langchain: 0.1.14 > langchain_community: 0.0.31 > langsmith: 0.1.24 > langchain_openai: 0.1.1 > langchain_text_splitters: 0.0.1 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve

HuggingFacePipeline does not use chat template

https://api.github.com/repos/langchain-ai/langchain/issues/19933/comments

2024-04-02T20:27:53Z

2024-08-06T22:56:36Z

https://github.com/langchain-ai/langchain/issues/19933

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code %pip install -U xmltodict from langchain_community.tools.pubmed.tool import PubmedQueryRun tool = PubmedQueryRun() tool.invoke("What causes lung cancer?") ### Error Message and Stack Trace (if applicable) 'PubMed exception: <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: self-signed certificate in certificate chain (_ssl.c:1006)>' ### Description I'm running your notebook and getting a certificate failure when hitting pubmed. https://github.com/langchain-ai/langchain/blob/master/docs/docs/integrations/tools/pubmed.ipynb ### System Info platform mac Python 3.11.7 Langchain Version: 0.1.14

Pubmed Tool SSL Cert Error when running code https://github.com/langchain-ai/langchain/blob/master/docs/docs/integrations/tools/pubmed.ipynb getting 'PubMed exception: <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: self-signed certificate in certificate chain (_ssl.c:1006)>'

https://api.github.com/repos/langchain-ai/langchain/issues/19916/comments

2024-04-02T12:47:20Z

2024-07-10T16:07:00Z

https://github.com/langchain-ai/langchain/issues/19916

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code > File "/usr/local/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py", line 319, in astream > async for chunk in _stream_implementation( > > > File "/usr/local/lib/python3.11/site-packages/langchain_openai/chat_models/base.py", line 511, in _astream > async for chunk in await self.async_client.create( > > > File "/usr/local/lib/python3.11/site-packages/openai/_streaming.py", line 150, in __aiter__ > async for item in self._iterator: > > > File "/usr/local/lib/python3.11/site-packages/openai/_streaming.py", line 181, in __stream__ > raise APIError( > ### Error Message and Stack Trace (if applicable) APIError('Failed to create completion as the model generated invalid Unicode output. Unfortunately, this can happen in rare situations. Consider reviewing your prompt or reducing the temperature of your request. You can retry your request, or contact us through our help center at help.openai.com if the error persists. (Please include the request ID req_b00ea6f9adc098af60178f40d8266c7d in your message.)')Traceback (most recent call last): File "/usr/local/lib/python3.11/site-packages/langchain_core/language_models/chat_models.py", line 319, in astream async for chunk in _stream_implementation( File "/usr/local/lib/python3.11/site-packages/langchain_openai/chat_models/base.py", line 511, in _astream async for chunk in await self.async_client.create( File "/usr/local/lib/python3.11/site-packages/openai/_streaming.py", line 150, in __aiter__ async for item in self._iterator: File "/usr/local/lib/python3.11/site-packages/openai/_streaming.py", line 181, in __stream__ raise APIError( ### Description Sometime there is no response ### System Info python3.11 ubuntu

OpenAI API: Failed to create completion as the model generated invalid Unicode output

https://api.github.com/repos/langchain-ai/langchain/issues/19914/comments

2024-04-02T12:05:19Z

2024-07-09T16:07:44Z

https://github.com/langchain-ai/langchain/issues/19914

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python _embedding = VertexAIEmbeddings(project="PROJECT_ID", credentials=gcp_auth.credentials, model_name="textembedding-gecko@003", location="PROJECT_LOCATION") _db = PGVector(connection_string="CONN", collection_name="COLLECTION_NAME", embedding_function=_embedding, use_jsonb=True) _db.search(query="test", search_type="mmr") ``` ### Error Message and Stack Trace (if applicable) ``` Traceback (most recent call last): File "/Users/amit/telly/telly-backend/backend/agent/healthcheck/components/hc_vector_db.py", line 29, in check_health db.search(query="test", search_type=self.settings.vector_db.retriever.type) File "/Users/amit/telly/telly-backend/backend/venv/lib/python3.12/site-packages/langchain_core/vectorstores.py", line 160, in search return self.max_marginal_relevance_search(query, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/amit/telly/telly-backend/backend/venv/lib/python3.12/site-packages/langchain_community/vectorstores/pgvector.py", line 1236, in max_marginal_relevance_search return self.max_marginal_relevance_search_by_vector( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/amit/telly/telly-backend/backend/venv/lib/python3.12/site-packages/langchain_community/vectorstores/pgvector.py", line 1314, in max_marginal_relevance_search_by_vector docs_and_scores = self.max_marginal_relevance_search_with_score_by_vector( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/amit/telly/telly-backend/backend/venv/lib/python3.12/site-packages/langchain_community/vectorstores/pgvector.py", line 1196, in max_marginal_relevance_search_with_score_by_vector mmr_selected = maximal_marginal_relevance( ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/amit/telly/telly-backend/backend/venv/lib/python3.12/site-packages/langchain_community/vectorstores/utils.py", line 34, in maximal_marginal_relevance similarity_to_query = cosine_similarity(query_embedding, embedding_list)[0] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/amit/telly/telly-backend/backend/venv/lib/python3.12/site-packages/langchain_community/utils/math.py", line 29, in cosine_similarity Z = 1 - simd.cdist(X, Y, metric="cosine") ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ TypeError: unsupported operand type(s) for -: 'int' and 'simsimd.DistancesTensor' INFO Cannot execute Healthcheck: hc-vector-db hc_vector_db.py:33 ``` ### Description Library Versions Used: * langchain~=0.1.14 * langchain-google-vertexai~=0.1.2 * simsimd~=4.2.2 ### System Info NA

TypeError: unsupported operand type(s) for -: 'int' and 'simsimd.DistancesTensor'

https://api.github.com/repos/langchain-ai/langchain/issues/19905/comments

2024-04-02T08:52:03Z

2024-07-16T16:06:48Z

https://github.com/langchain-ai/langchain/issues/19905

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code from langchain.agents.react.agent import create_react_agent from langchain.tools.retriever import create_retriever_tool from langchain.vectorstores.chroma import Chroma from langchain_openai import OpenAI retr_tool = Chroma().as_retriever() retriever_tool = create_retriever_tool(retr_tool, "db_search", "Searches and returns excerpts from database.") tools = [retriever_tool] llm = OpenAI(temperature=0) prompt = PromptTemplate( input_variables=["input", "agent_scratchpad"], template=""" --- REACT PROMPT HERE --- Query:... Thought: ... Action: ... Action Input: ... Observation: ... ... (this Thought/Action/Action Input/Observation can repeat N times) Thought: I am finished executing the plan (or, I cannot finish executing the plan without knowing some other information.) Final Answer: the final output from executing the plan or missing information I'd need to re-plan correctly. Query: {input}. Thought: {agent_scratchpad}""" ) # create the Agent react_runnable = create_react_agent( tools=tools, llm=llm, prompt=prompt, ) ### Error Message and Stack Trace (if applicable) _No response_ ### Description I'm trying to use create_react_agent to create an agent for an AgentExecutor, as ZeroShotAgent is marked as deprecated and will be removed in 0.2.0. The problem is that create_react_agent creates a Runnable, and not a Union[BaseSingleActionAgent, BaseMultiActionAgent], so it can't be used as input of AgentExecutor, as suggested in different parts of the documentation (see [here](https://python.langchain.com/docs/modules/agents/agent_types/react#create-agent)). ### System Info System Information ------------------ > OS: Linux > OS Version: #26~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Mar 12 10:22:43 UTC 2 > Python Version: 3.12.2 (main, Feb 25 2024, 16:35:05) [GCC 11.4.0] Package Information ------------------- > langchain_core: 0.1.36 > langchain: 0.1.13 > langchain_community: 0.0.29 > langsmith: 0.1.37 > langchain_openai: 0.0.8 > langchain_text_splitters: 0.0.1 > langchainhub: 0.1.15 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langgraph > langserve

create_react_agent cannot replace ZeroShotAgent

https://api.github.com/repos/langchain-ai/langchain/issues/19904/comments

2024-04-02T07:56:49Z

2024-07-11T16:07:09Z

https://github.com/langchain-ai/langchain/issues/19904

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code def _create_chat_result(self, response: Union[dict, BaseModel]) -> ChatResult: generations = [] if not isinstance(response, dict): response = response.dict() for res in response["choices"]: message = convert_dict_to_message(res["message"]) gen = ChatGeneration( message=message, generation_info=dict(finish_reason=res.get("finish_reason")), ) generations.append(gen) token_usage = response.get("usage", {}) llm_output = { "token_usage": token_usage, "model_name": self.model_name, "system_fingerprint": response.get("system_fingerprint", ""), } return ChatResult(generations=generations, llm_output=llm_output) ### Error Message and Stack Trace (if applicable) File "/Users/yuelu/anaconda3/envs/localgpt2/lib/python3.10/site-packages/langchain/chains/base.py", line 381, in acall raise e File "/Users/yuelu/anaconda3/envs/localgpt2/lib/python3.10/site-packages/langchain/chains/base.py", line 375, in acall await self._acall(inputs, run_manager=run_manager) | | | -> <langchain_core.callbacks.manager.AsyncCallbackManagerForChainRun object at 0x297397ee0> | | -> {'context': '文档内容总结', 'input': 'Python९ҕॉ൭Ҩ (PythonLibrary\nReference)\nRelease2.3.3\nGuidovanRossum\nFredL. Drake,Jr.,editor... | -> <function LLMChain._acall at 0x138bb9630> -> LLMChain(verbose=True, prompt=PromptTemplate(input_variables=['context', 'input'], template='你非常擅长总结文章，请根据{context}，总结文章内容。\n... File "/Users/yuelu/anaconda3/envs/localgpt2/lib/python3.10/site-packages/langchain/chains/llm.py", line 276, in _acall response = await self.agenerate(, run_manager=run_manager) | | | -> <langchain_core.callbacks.manager.AsyncCallbackManagerForChainRun object at 0x297397ee0> | | -> {'context': '文档内容总结', 'input': 'Python९ҕॉ൭Ҩ (PythonLibrary\nReference)\nRelease2.3.3\nGuidovanRossum\nFredL. Drake,Jr.,editor... | -> <function LLMChain.agenerate at 0x138bb9240> -> LLMChain(verbose=True, prompt=PromptTemplate(input_variables=['context', 'input'], template='你非常擅长总结文章，请根据{context}，总结文章内容。\n... File "/Users/yuelu/anaconda3/envs/localgpt2/lib/python3.10/site-packages/langchain/chains/llm.py", line 143, in agenerate return await self.llm.agenerate_prompt( | | -> <function BaseChatModel.agenerate_prompt at 0x1385db6d0> | -> ChatOpenAI(client=<class 'openai.api_resources.chat_completion.ChatCompletion'>, openai_api_key='3e7fdf2aaf9da189922784c997a9... -> LLMChain(verbose=True, prompt=PromptTemplate(input_variables=['context', 'input'], template='你非常擅长总结文章，请根据{context}，总结文章内容。\n... File "/Users/yuelu/anaconda3/envs/localgpt2/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py", line 501, in agenerate_prompt return await self.agenerate( | -> <function BaseChatModel.agenerate at 0x1385db5b0> -> ChatOpenAI(client=<class 'openai.api_resources.chat_completion.ChatCompletion'>, openai_api_key='3e7fdf2aaf9da189922784c997a9... File "/Users/yuelu/anaconda3/envs/localgpt2/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py", line 461, in agenerate raise exceptions[0] -> [KeyError('choices')] File "/Users/yuelu/anaconda3/envs/localgpt2/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py", line 577, in _agenerate_with_cache result = await self._agenerate( | -> <function ChatOpenAI._agenerate at 0x13fe9e5f0> -> ChatOpenAI(client=<class 'openai.api_resources.chat_completion.ChatCompletion'>, openai_api_key='3e7fdf2aaf9da189922784c997a9... File "/Users/yuelu/anaconda3/envs/localgpt2/lib/python3.10/site-packages/langchain_community/chat_models/openai.py", line 525, in _agenerate return self._create_chat_result(response) | | -> <OpenAIObject at 0x28ebe8720> JSON: { | | "error": { | | "message": "This model's maximum context length is 16385 tokens. Howev... | -> <function ChatOpenAI._create_chat_result at 0x13fe9e4d0> -> ChatOpenAI(client=<class 'openai.api_resources.chat_completion.ChatCompletion'>, openai_api_key='3e7fdf2aaf9da189922784c997a9... File "/Users/yuelu/anaconda3/envs/localgpt2/lib/python3.10/site-packages/langchain_community/chat_models/openai.py", line 454, in _create_chat_result for res in response["choices"]: -> <OpenAIObject at 0x28ebe8720> JSON: { "error": { "message": "This model's maximum context length is 16385 tokens. Howev... KeyError: 'choices' ### Description I'm trying to use langchain to summarize the content of the document the document is very long, so the llm's response is "error": { "message": "This model's maximum context length is 16385 tokens.xxx" } but langchain_community.chat_models.openai.ChatOpenAI._create_chat_result hides the original response and throw the exception KeyError: 'choices' i want to get the original error ### System Info langchain 0.1.0 python 3.10 mac and linux

langchain_community.chat_models.openai.ChatOpenAI._create_chat_result hides the original response

https://api.github.com/repos/langchain-ai/langchain/issues/19898/comments

2024-04-02T03:42:39Z

2024-07-10T16:06:46Z

https://github.com/langchain-ai/langchain/issues/19898

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code The error is in the filer apache_doris.py, folder vectorstores. The following code ``` else f"array<float>{str(_n)}" ```, need to be chage to ``` else f"{str(_n)}" ``` ### Error Message and Stack Trace (if applicable) Error <class 'pymysql.err.OperationalError'> (1105, "ParseException, ms.... ### Description If we remove the sring "array<float>" it will work. I have tested with apache doris 2.1.0 ### System Info Apache doris 2.1.0 Last version of langchain

Error in apache doris insert

https://api.github.com/repos/langchain-ai/langchain/issues/19886/comments

2024-04-01T21:00:23Z

2024-07-11T16:07:04Z

https://github.com/langchain-ai/langchain/issues/19886

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code python... here is the code that i use: i ahve index data on azure cognitive search and each chunk has searchable type of metadata which is location if i use the use `acs.as_retriever()` function along with filter it retrieve different location data as well like you can see in below code output is have print each retrieve doc location metadata. but i use the `acs.similarity_search() ` and inside we pass filter it will only retrieve that location data and not retrieve mix location data. ``` acs = acs_search("testindex") retriever = acs.as_retriever(search_kwargs={'filter': {'location':'US'}, 'k': 5}) def format_docs(docs): for i in docs: print(i.metadata["location"]) return "\n\n".join(doc.page_content for doc in docs) rag_chain = ( {"context": retriever | format_docs, "question": RunnablePassthrough()} | prompt | llm | StrOutputParser() ) rag_chain.invoke("What is hr policy about leave") ``` **OUTPUT** ``` US PK MY US MY 'The HR policy about leave at xyz includes standard paid leave for full-time employees after 90 days of continuous employment. This includes Annual Leave (AL) of 14 workdays.... ``` ### USE acs.similarity_search() ``` res = acs.similarity_search( query="what is the hr policy for anual leave", k=4, search_type="hybrid", filters="location eq 'US'" ) res ``` OUTPUT: ``` [Document(page_content='Leave taken under this policy does, metadata={'source': '2023-us.pdf', 'location': 'US'}), Document(page_content='You may use available vacation, pers metadata={'source': '2023-us.pdf', 'location': 'US'}), Document(page_content="Failure to Return to Work If you fa", metadata={'source': '2023-us.pdf', 'location': 'US'}), Document(page_content='To request leave under this policy, , metadata={'source': '2023-us.pdf', 'location': 'US'})] ``` you can see this function give exact filter data and not mixed data . what would be the solution because we are use the first function inside chain and we are unable to get filter data. ### Error Message and Stack Trace (if applicable) inside langchain_core > vectorstores.py i have place this print but the filter did not work: ``` def _get_relevant_documents( self, query: str, *, run_manager: CallbackManagerForRetrieverRun ) -> List[Document]: if self.search_type == "similarity": print("===filter=======>\n",self.search_kwargs,"\n=============") docs = self.vectorstore.similarity_search(query, **self.search_kwargs) ``` OUTPUT: ``` ===filter=======> {'filter': {'location': 'US'}, 'k': 5} ============= ``` we are unable to get filter data while using `as_retriever()` function inside chain the doc return by this is given in first code output ### Description i use the below versions ``` langchain==0.1.8 langchain-community==0.0.21 langchain-core==0.1.25 langchain-openai==0.0.6 ``` ### System Info ``` aiohttp==3.9.3 aiosignal==1.3.1 annotated-types==0.6.0 antlr4-python3-runtime==4.9.3 anyio==4.3.0 argon2-cffi==23.1.0 argon2-cffi-bindings==21.2.0 arrow==1.3.0 asgiref==3.7.2 asttokens==2.4.1 async-lru==2.0.4 async-timeout==4.0.3 asyncio-redis==0.16.0 attrs==23.2.0 azure-common==1.1.28 azure-core==1.30.0 azure-identity==1.15.0 azure-monitor-opentelemetry-exporter==1.0.0b22 azure-search-documents==11.4.0 azure-storage-blob==12.19.1 Babel==2.14.0 backoff==2.2.1 beautifulsoup4==4.12.3 bleach==6.1.0 cachetools==5.3.3 certifi==2024.2.2 cffi==1.16.0 chardet==5.2.0 charset-normalizer==3.3.2 click==8.1.7 cohere==4.56 coloredlogs==15.0.1 comm==0.2.1 contourpy==1.2.0 cryptography==42.0.4 cycler==0.12.1 dataclasses-json==0.6.4 debugpy==1.8.1 decorator==5.1.1 deepdiff==6.7.1 defusedxml==0.7.1 Deprecated==1.2.14 distro==1.9.0 effdet==0.4.1 emoji==2.10.1 et-xmlfile==1.1.0 exceptiongroup==1.2.0 executing==2.0.1 fastapi==0.109.2 fastavro==1.9.4 fastjsonschema==2.19.1 filelock==3.13.1 filetype==1.2.0 fixedint==0.1.6 flatbuffers==24.3.6 fonttools==4.49.0 fqdn==1.5.1 frozenlist==1.4.1 fsspec==2024.2.0 greenlet==3.0.3 h11==0.14.0 httpcore==1.0.4 httpx==0.27.0 huggingface-hub==0.21.4 humanfriendly==10.0 idna==3.6 importlib-metadata==6.11.0 iopath==0.1.10 ipykernel==6.29.2 ipython==8.22.1 ipywidgets==8.1.2 isodate==0.6.1 isoduration==20.11.0 jedi==0.19.1 Jinja2==3.1.3 joblib==1.3.2 json5==0.9.24 jsonpatch==1.33 jsonpath-python==1.0.6 jsonpointer==2.4 jsonschema==4.21.1 jsonschema-specifications==2023.12.1 jupyter==1.0.0 jupyter-console==6.6.3 jupyter-events==0.10.0 jupyter-lsp==2.2.4 jupyter_client==8.6.0 jupyter_core==5.7.1 jupyter_server==2.13.0 jupyter_server_terminals==0.5.3 jupyterlab==4.1.5 jupyterlab_pygments==0.3.0 jupyterlab_server==2.25.4 jupyterlab_widgets==3.0.10 kiwisolver==1.4.5 langchain==0.1.8 langchain-community==0.0.21 langchain-core==0.1.25 langchain-openai==0.0.6 langchainhub==0.1.15 langdetect==1.0.9 langsmith==0.1.5 layoutparser==0.3.4 lxml==5.1.0 MarkupSafe==2.1.5 marshmallow==3.20.2 matplotlib==3.8.3 matplotlib-inline==0.1.6 mistune==3.0.2 mpmath==1.3.0 msal==1.26.0 msal-extensions==1.1.0 msrest==0.7.1 multidict==6.0.5 mypy-extensions==1.0.0 nbclient==0.10.0 nbconvert==7.16.3 nbformat==5.10.3 nest-asyncio==1.6.0 networkx==3.2.1 nltk==3.8.1 notebook==7.1.2 notebook_shim==0.2.4 numpy==1.26.4 nvidia-cublas-cu12==12.1.3.1 nvidia-cuda-cupti-cu12==12.1.105 nvidia-cuda-nvrtc-cu12==12.1.105 nvidia-cuda-runtime-cu12==12.1.105 nvidia-cudnn-cu12==8.9.2.26 nvidia-cufft-cu12==11.0.2.54 nvidia-curand-cu12==10.3.2.106 nvidia-cusolver-cu12==11.4.5.107 nvidia-cusparse-cu12==12.1.0.106 nvidia-nccl-cu12==2.19.3 nvidia-nvjitlink-cu12==12.4.99 nvidia-nvtx-cu12==12.1.105 oauthlib==3.2.2 omegaconf==2.3.0 onnx==1.15.0 onnxruntime==1.15.1 openai==1.12.0 opencv-python==4.9.0.80 openpyxl==3.1.2 opentelemetry-api==1.22.0 opentelemetry-instrumentation==0.43b0 opentelemetry-instrumentation-asgi==0.43b0 opentelemetry-instrumentation-fastapi==0.43b0 opentelemetry-sdk==1.22.0 opentelemetry-semantic-conventions==0.43b0 opentelemetry-util-http==0.43b0 ordered-set==4.1.0 overrides==7.7.0 packaging==23.2 pandas==2.2.1 pandocfilters==1.5.1 parso==0.8.3 pdf2image==1.17.0 pdfminer.six==20221105 pdfplumber==0.10.4 pexpect==4.9.0 pikepdf==8.13.0 pillow==10.2.0 pillow_heif==0.15.0 platformdirs==4.2.0 portalocker==2.8.2 prometheus_client==0.20.0 prompt-toolkit==3.0.43 protobuf==4.25.3 psutil==5.9.8 ptyprocess==0.7.0 pure-eval==0.2.2 pycocotools==2.0.7 pycparser==2.21 pydantic==2.6.1 pydantic-settings==2.2.0 pydantic_core==2.16.2 Pygments==2.17.2 PyJWT==2.8.0 pymssql==2.2.11 pyparsing==3.1.2 pypdf==4.1.0 pypdfium2==4.27.0 pytesseract==0.3.10 python-dateutil==2.8.2 python-docx==1.1.0 python-dotenv==1.0.1 python-iso639==2024.2.7 python-json-logger==2.0.7 python-magic==0.4.27 python-multipart==0.0.9 pytz==2024.1 PyYAML==6.0.1 pyzmq==25.1.2 qtconsole==5.5.1 QtPy==2.4.1 rapidfuzz==3.6.2 redis==5.0.1 referencing==0.34.0 regex==2023.12.25 requests==2.31.0 requests-oauthlib==1.3.1 rfc3339-validator==0.1.4 rfc3986-validator==0.1.1 rpds-py==0.18.0 safetensors==0.4.2 scipy==1.12.0 Send2Trash==1.8.2 six==1.16.0 sniffio==1.3.0 soupsieve==2.5 SQLAlchemy==2.0.27 stack-data==0.6.3 starlette==0.36.3 sympy==1.12 tabulate==0.9.0 tenacity==8.2.3 terminado==0.18.1 tiktoken==0.6.0 timm==0.9.16 tinycss2==1.2.1 tokenizers==0.15.2 tomli==2.0.1 torch==2.2.1 torchvision==0.17.1 tornado==6.4 tqdm==4.66.2 traitlets==5.14.1 transformers==4.38.2 triton==2.2.0 types-python-dateutil==2.9.0.20240316 types-requests==2.31.0.20240311 typing-inspect==0.9.0 typing_extensions==4.9.0 tzdata==2024.1 unstructured==0.12.4 unstructured-client==0.21.1 unstructured-inference==0.7.23 unstructured.pytesseract==0.3.12 uri-template==1.3.0 urllib3==2.2.1 uvicorn==0.27.1 wcwidth==0.2.13 webcolors==1.13 webencodings==0.5.1 websocket-client==1.7.0 widgetsnbextension==4.0.10 wrapt==1.16.0 xlrd==2.0.1 yarl==1.9.4 zipp==3.17.0 ```

retriever.as_retriever() function cannot retrieve data based on filter from azure cognitive search but

https://api.github.com/repos/langchain-ai/langchain/issues/19885/comments

2024-04-01T20:24:40Z

2024-07-11T16:06:59Z

https://github.com/langchain-ai/langchain/issues/19885

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code https://js.langchain.com/docs/modules/chains/popular/vector_db_qa ### Error Messaged and Stack Trace (if applicable) _No response_ ### Description Several platforms offer the OpenAI API as an alternative to their own API. Examples include TGI and vLLM. However, in these platforms, prompts must be provided as a single string rather than a list of strings. Unfortunately, when using the QA retrieval chain, I encounter an error because it requires a list of prompts. Issues related to this: https://github.com/vllm-project/vllm/issues/186 https://github.com/vllm-project/vllm/pull/323#issuecomment-1615254198 ### System Info `langchain 0.1.14 ` Windows

Langchain passes prompt as a list instead of str in Retrieval Chain to Openai API

https://api.github.com/repos/langchain-ai/langchain/issues/19872/comments

2024-04-01T17:38:06Z

2024-07-09T16:07:19Z

https://github.com/langchain-ai/langchain/issues/19872

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code import getpass import os from langchain_core.pydantic_v1 import BaseModel, Field # Note that the docstrings here are crucial, as they will be passed along # to the model along with the class name. class Multiply(BaseModel): """Multiply two integers together.""" a: int = Field(..., description="First integer") b: int = Field(..., description="Second integer") os.environ["GOOGLE_API_KEY"] = getpass.getpass() from langchain_google_genai import ChatGoogleGenerativeAI llm = ChatGoogleGenerativeAI(model="gemini-pro") llm_with_tools = llm.bind_tools([Multiply]) llm_with_tools.invoke("what's 3 * 12") ### Error Message and Stack Trace (if applicable) langchaingemsample.py", line 20, in <module> llm_with_tools = llm.bind_tools([Multiply]) ^^^^^^^^^^^^^^ AttributeError: 'ChatGoogleGenerativeAI' object has no attribute 'bind_tools' ### Description i was using using langchain examples in this url https://python.langchain.com/docs/modules/model_io/chat/function_calling on function calling using google, i got this error "AttributeError: 'ChatGoogleGenerativeAI' object has no attribute 'bind_tools'", what i'm i doing wrong? ### System Info System Information ------------------ > OS: Windows > OS Version: 10.0.19045 > Python Version: 3.11.4 (tags/v3.11.4:d2340ef, Jun 7 2023, 05:45:37) [MSC v.1934 64 bit (AMD64)] Package Information ------------------- > langchain_core: 0.1.37 > langchain: 0.1.13 > langchain_community: 0.0.29 > langsmith: 0.1.31 > langchain_experimental: 0.0.50 > langchain_google_genai: 1.0.1 > langchain_google_vertexai: 0.1.1 > langchain_openai: 0.1.1 > langchain_text_splitters: 0.0.1 > langchaingemsample: Installed. No version info available. > langchainhub: 0.1.14 > langgraph: 0.0.30 Packages not installed (Not Necessarily a Problem) -------------------------------------------------- The following packages were not found: > langserve

AttributeError: 'ChatGoogleGenerativeAI' object has no attribute 'bind_tools'

https://api.github.com/repos/langchain-ai/langchain/issues/19868/comments

2024-04-01T17:00:09Z

2024-07-09T16:07:15Z

https://github.com/langchain-ai/langchain/issues/19868

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ```python from langchain.chats import ChatOpenAI, ChatPromptTemplate, SystemMessagePromptTemplate, HumanMessagePromptTemplate endpoint_url = "http://url/v1" chat = ChatOpenAI( model="mistral-medium", openai_api_key="EMPTY", openai_api_base=endpoint_url, max_tokens=5, temperature=0, ) template = ( "You are a helpful assistant that translates {input_language} to {output_language}." ) system_message_prompt = SystemMessagePromptTemplate.from_template(template) human_template = "{text}" human_message_prompt = HumanMessagePromptTemplate.from_template(human_template) chat_prompt = ChatPromptTemplate.from_messages( [system_message_prompt, human_message_prompt] ) results = chat( chat_prompt.format_prompt( input_language="English", output_language="Italian", text="I love programming." ).to_messages() ) print(results) ``` ### Error Message and Stack Trace (if applicable) ```python openai.UnprocessableEntityError: Error code: 422 - {'error': 'Template error: template not found', 'error_type': 'template_error'} ``` ### Description I'm encountering an issue with LangChain's OpenAI Chat integration, specifically when trying to use custom templates for dynamic chat prompts. The expected behavior is to seamlessly generate chat prompts using the specified system and human message templates, and then perform a chat session using these prompts. However, the current behavior results in an **openai.UnprocessableEntityError** with a **template error** indicating that the template was not found. This issue arises despite following the documented approach for creating and using **ChatPromptTemplate**, **SystemMessagePromptTemplate**, and **HumanMessagePromptTemplate** within LangChain's framework. The error suggests there's either a problem with the template processing/handling in LangChain or a misconfiguration in the OpenAI chat integration setup. Even when using a cURL command to interact with the model, it's still resulting in an error about a missing template. #### cURL Testing ```curl curl http://ur/lv1/chat/completions \ -H "Content-Type: application/json" \ -d '{ "model": "mistralai/Mistral-7B-Instruct-v0.2", "messages": [ { "role": "system", "content": "When user gives a number, simply respond with the double and say nothing else." }, { "role": "user", "content": "100" }] }' ``` ### System Info System Information ------------------ > OS: Linux > OS Version: #224-Ubuntu SMP Mon Jun 19 13:30:12 UTC 2023 > Python Version: 3.10.0 (default, Jan 10 2024, 16:43:55) [GCC 7.5.0] Package Information ------------------- > langchain_core: 0.1.37 > langchain: 0.1.14 > langchain_community: 0.0.30 > langsmith: 0.1.38 > langchain_mistralai: 0.1.0 > langchain_openai: 0.1.1 > langchain_text_splitters: 0.0.1

Bug Report: Issue with LangChain and OpenAI Template Handling (vLLM)

https://api.github.com/repos/langchain-ai/langchain/issues/19857/comments

2024-04-01T13:56:20Z

2024-04-10T08:09:25Z

https://github.com/langchain-ai/langchain/issues/19857

### Checked other resources - [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package). ### Example Code ~~~python # Python code from langchain_openai import OpenAIEmbeddings from langchain_community.vectorstores import Chroma vectorstore = Chroma.from_documents(documents=splits, embedding=OpenAIEmbeddings(model="text-embedding-3-large"), persist_directory = "./chroma_db") ~~~ ### Error Message and Stack Trace (if applicable) File /workspace/.pyenv_mirror/user/current/lib/python3.12/site-packages/chromadb/api/types.py:211, in validate_embedding_function(embedding_function) [208](https://vscode-remote+utkarshg1-002dopenailangcha-002dh4687x6hb59-002ews-002dus110-002egitpod-002eio.vscode-resource.gitpod.io/workspace/.pyenv_mirror/user/current/lib/python3.12/site-packages/chromadb/api/types.py:208) protocol_signature = signature(EmbeddingFunction.__call__).parameters.keys() [210](https://vscode-remote+utkarshg1-002dopenailangcha-002dh4687x6hb59-002ews-002dus110-002egitpod-002eio.vscode-resource.gitpod.io/workspace/.pyenv_mirror/user/current/lib/python3.12/site-packages/chromadb/api/types.py:210) if not function_signature == protocol_signature: --> [211](https://vscode-remote+utkarshg1-002dopenailangcha-002dh4687x6hb59-002ews-002dus110-002egitpod-002eio.vscode-resource.gitpod.io/workspace/.pyenv_mirror/user/current/lib/python3.12/site-packages/chromadb/api/types.py:211) raise ValueError( [212](https://vscode-remote+utkarshg1-002dopenailangcha-002dh4687x6hb59-002ews-002dus110-002egitpod-002eio.vscode-resource.gitpod.io/workspace/.pyenv_mirror/user/current/lib/python3.12/site-packages/chromadb/api/types.py:212) f"Expected EmbeddingFunction.__call__ to have the following signature: {protocol_signature}, got {function_signature}\n" [213](https://vscode-remote+utkarshg1-002dopenailangcha-002dh4687x6hb59-002ews-002dus110-002egitpod-002eio.vscode-resource.gitpod.io/workspace/.pyenv_mirror/user/current/lib/python3.12/site-packages/chromadb/api/types.py:213) "Please see https://docs.trychroma.com/embeddings for details of the EmbeddingFunction interface.\n" [214](https://vscode-remote+utkarshg1-002dopenailangcha-002dh4687x6hb59-002ews-002dus110-002egitpod-002eio.vscode-resource.gitpod.io/workspace/.pyenv_mirror/user/current/lib/python3.12/site-packages/chromadb/api/types.py:214) "Please note the recent change to the EmbeddingFunction interface: https://docs.trychroma.com/migration#migration-to-0416---november-7-2023 \n" [215](https://vscode-remote+utkarshg1-002dopenailangcha-002dh4687x6hb59-002ews-002dus110-002egitpod-002eio.vscode-resource.gitpod.io/workspace/.pyenv_mirror/user/current/lib/python3.12/site-packages/chromadb/api/types.py:215) ) ValueError: Expected EmbeddingFunction.__call__ to have the following signature: odict_keys(['self', 'input']), got odict_keys(['args', 'kwargs']) Please see https://docs.trychroma.com/embeddings for details of the EmbeddingFunction interface. Please note the recent change to the EmbeddingFunction interface: https://docs.trychroma.com/migration#migration-to-0416---november-7-2023 ### Description I'm Trying to get Embeddings but looks like ChromaDB has changed functionality from their end. Hence Langchain code also needs to be changed https://docs.trychroma.com/embeddings/openai ### System Info langchain==0.1.14 langchain-community==0.0.30 langchain-core==0.1.37 langchain-openai==0.1.1 langchain-text-splitters==0.0.1 langchainhub==0.1.15 chromadb==0.4.24

Chroma db Code changed thats why unable to access the vectorstore from ChromaDB for embeddings

https://api.github.com/repos/langchain-ai/langchain/issues/19848/comments

2024-04-01T07:33:39Z

2024-04-01T17:56:47Z

https://github.com/langchain-ai/langchain/issues/19848