anybody experiencing error while converting pdf to text in the shown method in the video, can use the following code to resolve: from PyPDF2 import PdfReader pdf_text = "" with open(pdf_file.path, "rb") as file: pdf_reader = PdfReader(file) for page in pdf_reader.pages: pdf_text += page.extract_text()
🎯 Key Takeaways for quick navigation: 00:00 📽️ Trong video này, tác giả giới thiệu về loạt video về Chainlit và kế hoạch cho nội dung các video trong loạt này. 01:08 🤖 Video sẽ sử dụng mô hình của OpenAI cho embeddings và mô hình ngôn ngữ lớn, cùng với Lang Chain, Chainlit, CromDB và Docker cho việc tạo ứng dụng và triển khai trên Google Cloud. 03:38 📚 Tác giả đã tạo sẵn các video giải thích chi tiết về cách tạo custom chatbot cho tài liệu và văn bản, cùng với các hướng dẫn về Chainlit và Lang Chain trong các video trước đó. Made with HARPA AI
getting this error - Traceback (most recent call last): File "D:\langchain-openai-chainlit-main\venv\lib\site-packages\chainlit\utils.py", line 39, in wrapper return await user_function(**params_values) File "D:\langchain-openai-chainlit-main\pdf_qa.py", line 82, in on_chat_start pdf_stream = BytesIO(file.content) AttributeError: 'AskFileResponse' object has no attribute 'content'
Hi, I went through, use codespace, the Chainlit starts fine, but when upload pdf always the following error: 'AskFileResponse' object has no attribute 'content'. Any tip, why is it? thx!
@@datasciencebasics thx, now I would like to see a working easy beginner friendly RAG app with gpt llm. can you suggest working script, tutorial, I guess tutorials from 6-8 month ago still not working now, the scripts are out of date now...
Hi, when I run your code on Codespace, the following error popped up: Document prompt requires documents to have metadata variables: ['source']. Received document with missing metadata: ['source']. Please help. By the way, thank you for this, very clear introduction.
Hello, I liked your content.. but it no longer works in 2024 because all libraries are outdated now, can you update to make it current. Just wanted to give you that feedback. Thanks
Hi , It is not working on follow up questions. Can you make video on same where we can ask follow question basis chat history using memory from langchain.TIA
@datasciencebasics I have tried to add different langchain.memory into RetrievalQAWithSourcesChain but it is failing to remember. Your help will be appreciated.
anybody experiencing error while converting pdf to text in the shown method in the video, can use the following code to resolve:
from PyPDF2 import PdfReader
pdf_text = ""
with open(pdf_file.path, "rb") as file:
pdf_reader = PdfReader(file)
for page in pdf_reader.pages:
pdf_text += page.extract_text()
Thanks for your input, appreciated 🙏🏼
@@datasciencebasics please pin it. It will help people stuck to find solution quickly
🎯 Key Takeaways for quick navigation:
00:00 📽️ Trong video này, tác giả giới thiệu về loạt video về Chainlit và kế hoạch cho nội dung các video trong loạt này.
01:08 🤖 Video sẽ sử dụng mô hình của OpenAI cho embeddings và mô hình ngôn ngữ lớn, cùng với Lang Chain, Chainlit, CromDB và Docker cho việc tạo ứng dụng và triển khai trên Google Cloud.
03:38 📚 Tác giả đã tạo sẵn các video giải thích chi tiết về cách tạo custom chatbot cho tài liệu và văn bản, cùng với các hướng dẫn về Chainlit và Lang Chain trong các video trước đó.
Made with HARPA AI
getting this error - Traceback (most recent call last):
File "D:\langchain-openai-chainlit-main\venv\lib\site-packages\chainlit\utils.py", line 39, in wrapper
return await user_function(**params_values)
File "D:\langchain-openai-chainlit-main\pdf_qa.py", line 82, in on_chat_start
pdf_stream = BytesIO(file.content)
AttributeError: 'AskFileResponse' object has no attribute 'content'
Gimana ya
Thanks for your explaination
You are welcome.
Thank you for this. Much appreciated.
You’re welcome :)
@datasciencebasics I am getting the error "TypeError: expected string or buffer". Kindly let me know how to resolve it
HAVE YOU GOT SOLUTION FOR THIS ERROR? I am getting same error @alislounge @datasciencebasics
@@gauravchaurasia6080
When I edit the code like this, it works.
`res = await chain.acall(message.content, callbacks=[cb])`
Same error...@@gauravchaurasia6080
Hi, I went through, use codespace, the Chainlit starts fine, but when upload pdf always the following error: 'AskFileResponse' object has no attribute 'content'. Any tip, why is it? thx!
Check the latest code here, Chat With Documents Using ChainLit, LangChain, Ollama & Mistral 🧠
th-cam.com/video/2IL0Sd3neWc/w-d-xo.html
@@datasciencebasics thx, but with Ollama can I use OpenAI LLL as well? or just opensource ones that I can download?
Ollama is for open source models
@@datasciencebasics thx, now I would like to see a working easy beginner friendly RAG app with gpt llm. can you suggest working script, tutorial, I guess tutorials from 6-8 month ago still not working now, the scripts are out of date now...
I get the same error !
Thank you Great content
You’re welcome !!
Can you make a video of performing authentication using chainlit
Hi, when I run your code on Codespace, the following error popped up: Document prompt requires documents to have metadata variables: ['source']. Received document with missing metadata: ['source']. Please help. By the way, thank you for this, very clear introduction.
Can you remove the codespace and run it again. I am not facing taht issue.
Hello, I liked your content.. but it no longer works in 2024 because all libraries are outdated now, can you update to make it current. Just wanted to give you that feedback. Thanks
Thanks for the suggestion. Will try my best to update as much as I can 🙏🏼
Hi , It is not working on follow up questions. Can you make video on same where we can ask follow question basis chat history using memory from langchain.TIA
Hello I have created other videos where I have explained how to add chat history. Please refer to that and plug it in Chainlit.
@datasciencebasics I have tried to add different langchain.memory into RetrievalQAWithSourcesChain but it is failing to remember. Your help will be appreciated.
I am getting the error "TypeError: expected string or buffer".
Hello, did u try the answer provided by another user in another comment.
You can resolve the error by replacing the contents of requirement.txt as below
langchain == 0.0.246
openai == 0.27.8
chainlit == 0.6.1
chromadb == 0.4.3
tiktoken == 0.4.0
PyPDF2 == 3.0.1
pypdf == 3.13.0
pandas == 2.0.3
tabulate == 0.9.0
no windows support hã?
It should work normally in Windows too. But, for easiness, You can use github codespace once I make the repo public.