Closed davidgross631 closed 1 month ago
贴一些日志吧,在文件夹 ragtest/output/202407xxxx/
里面
Attach some logs, which can be found in ragtest/output/202407xxxx/
folder.
Not sure how helpful this is, but thank you regardless!
15:33:28,850 graphrag.index.input.text INFO found text files from input, found [('book.txt', {})]
15:33:28,855 graphrag.index.input.text WARNING Warning! Error loading file book.txt. Skipping...
15:33:28,855 graphrag.index.input.text INFO Found 1 files, loading 0
15:33:28,855 graphrag.index.workflows.load INFO Workflow Run Order: ['create_base_text_units', 'create_base_extracted_entities', 'create_summarized_entities', 'create_base_entity_graph', 'create_final_entities', 'create_final_nodes', 'create_final_communities', 'join_text_units_to_entity_ids', 'create_final_relationships', 'join_text_units_to_relationship_ids', 'create_final_community_reports', 'create_final_text_units', 'create_base_documents', 'create_final_documents']
create_base_text_units
workflow got error, caused by 'id'. First, Check the boot.txt
content. and then check your prompt.
book.txt
This is the book.txt content, not sure why there would be an issue because I followed this command from the guide online:
curl https://www.gutenberg.org/cache/epub/24022/pg24022.txt > ./ragtest/input/book.txt
Maybe, I just put an excerpt from the book in book.txt and move on from there?
Now, I am getting this error in the log: logs.json
@davidgross631 can you try setting encoding_model: o200k_base
in your settings.yaml? With gpt4o OAI changed their encoding model (see tiktoken mapping). We can update the docs to help clarify this.
Hm, still having issues unfortunately. Very similar to the Error #779
I do not have/see a indexing-engine.log but I commented out the api_base line in settings.yaml to put in https://api.openai.com/v1/
Still did not work sadly. Still getting this same message. I also tried to mess with the max-tokens as reffered to in a different issue but still no luck.
Ok, thanks for trying that. So your issue may be happening before the encoding_model setting is even a potential problem then. Are you directly on Windows, or using WSL? I am hearing there can be utf-8 encoding issues with default Windows, and folks are having luck running in WSL if that's an option to try.
book.txt This is the book.txt content, not sure why there would be an issue because I followed this command from the guide online:
curl https://www.gutenberg.org/cache/epub/24022/pg24022.txt > ./ragtest/input/book.txt
Maybe, I just put an excerpt from the book in book.txt and move on from there?
Certainly worth a try in case curl didn't download it correctly
Windows, I've already set the text to UTF-8 encoding using Notepad++
I have also used WSL/Ubuntu and Windows with no luck for either
I have also used WSL/Ubuntu and Windows with no luck for either
After deleting cache folder and output folder, I got same error every tries.
Hm, still having issues unfortunately. Very similar to the Error #779
I do not have/see a indexing-engine.log but I commented out the api_base line in settings.yaml to put in https://api.openai.com/v1/
Still did not work sadly. Still getting this same message. I also tried to mess with the max-tokens as reffered to in a different issue but still no luck.
I have the same problem. Have you solved it?
Hm, still having issues unfortunately. Very similar to the Error #779 I do not have/see a indexing-engine.log but I commented out the api_base line in settings.yaml to put in https://api.openai.com/v1/ Still did not work sadly. Still getting this same message. I also tried to mess with the max-tokens as reffered to in a different issue but still no luck.
I have the same problem. Have you solved it?
I have, it was actually an issue with the OpenAI API key. Make sure that you are not using a free version or atleast double check that there aren't any billing/rate issues. That was the case with me and didn't realize it until trying it with WSL.
I'm still facing this exact same issue. I am using Azure Open AI instead of Open AI. Anyone who was using Azure OpenAI that has solved this? Performed all the recommended steps but no luck.
Is there an existing issue for this?
Describe the issue
Ok, so basically I'm following this set-up guide:
Get Started (microsoft.github.io)
Everything was going fine until I ran the pipeline with this command:
python -m graphrag.index --root ./ragtest
This is the error:
I am using Open AI and I have the proper API key put in .env file. However, I did change the settings.yaml file to have the model be gpt-4o-mini because my API key supports that.
I have tried looking at the log files and no real help there either. I am just wondering what the issue could possibly be caused by.
Steps to reproduce
You can replicate the issue basically by following the exact same steps here and use OpenAI NOT AzureOpenAI:
https://microsoft.github.io/graphrag/posts/get_started/
GraphRAG Config Used
Logs and screenshots
Log Folder:
Main.log:
Network.log in window2:
Additional Information