samwit / langchain-tutorials Goto Github PK

I went through “langchain-tutorials” but cannot find the code in the video"LangChain + Retrieval Local LLMs for Retrieval QA - No OpenAI!!!" . Can you help me? Thanks

Hi Sam,

In the conversation with memory tutorial, you have the lines:

from pprint import pprint
pprint(conversation.memory.store)

store apparently is no longer an attribute of memory. What would be the proper replacement? thanks. I get the error:

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
[/Users/erlebach/src/2023/langchain_ecosystem/langchain_tutorials_wittenveek/YT_LangChain_Basic_Conversation_Chatbot_with_Memory_Demo.ipynb](https://file+.vscode-resource.vscode-cdn.net/Users/erlebach/src/2023/langchain_ecosystem/langchain_tutorials_wittenveek/YT_LangChain_Basic_Conversation_Chatbot_with_Memory_Demo.ipynb) Cell 67 line 2
      [1](vscode-notebook-cell:/Users/erlebach/src/2023/langchain_ecosystem/langchain_tutorials_wittenveek/YT_LangChain_Basic_Conversation_Chatbot_with_Memory_Demo.ipynb#Y123sZmlsZQ%3D%3D?line=0) from pprint import pprint
----> [2](vscode-notebook-cell:/Users/erlebach/src/2023/langchain_ecosystem/langchain_tutorials_wittenveek/YT_LangChain_Basic_Conversation_Chatbot_with_Memory_Demo.ipynb#Y123sZmlsZQ%3D%3D?line=1) pprint(conversation.memory.store)

AttributeError: 'ConversationEntityMemory' object has no attribute 'store'

Thanks! Great tutorial videos by the way!

Gordon

How to improve agents response in terms of borderless tables? A table where there are no definite headers, these tables are present in 10k reports, when i extract them from pdfs and store them in csv's i want to ask questions with them or say such a table is an example, i see the question answering isnt that great, i have received indefinite answers, or the GPT has denied an answer, did you work on such scenarios?

Hi,

First of all, thank you for sharing all this on YouTube.
I ran the script rag.py exactly the way you did and I am getting the following issue, could you please run your script back and see if it breaks as well?

ValueError: Expected EmbeddingFunction.call to have the following signature: odict_keys(['self', 'input']), got odict_keys(['self', '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

the
from langchain.llms import Ollama
has to be replaced with
from langchain_community.llms import Ollama

Thank you for making good videos. I was wondering how to find the related examples, especially about the 16K tokens LangChain example, if you have published it? Sorry if this is a stupid question, but I didn't notice it here.