New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Question classifier proves to be ineffective #4273
Comments
To address the ineffectiveness of the question classifier in Dify version 0.6.7, consider the following targeted strategies:
Ensure your Docker environment is correctly set up and all dependencies are properly installed. If issues persist, providing specific examples of misclassified questions or any error messages could help in diagnosing and resolving the problem more effectively.
|
What LLM do you use? A lot of it has to do with LLM performance. You can provide more information. |
Can you provide more evidence for the community to further investgate this issue please? |
Deploying the Qwen 14b model locally on Xinference. If I use qwen-max-0403,everything looks ok |
Back to 0.6.6 version, the Question classifier still be ineffective |
@tigerun I think it has to do with the LLM capability and the service is fine |
0.6.6 everything is ok,but when update to 0.6.7 ,Question classifier look like a idiot,do nothing except update. |
To resolve the issue with the question classifier version 0.6.7, follow these steps: Locate services.api.image in your docker-compose.yaml file. |
Self Checks
Dify version
0.6.7
Cloud or Self Hosted
Self Hosted (Docker)
Steps to reproduce
After many times of attempts, it is evident that the 0.6.7 version of the question classifier abysmal intellect for problem categorization, frequently failing to identify issues accurately.
✔️ Expected Behavior
Urgent rectification
❌ Actual Behavior
Question classifier proves to be ineffective
The text was updated successfully, but these errors were encountered: