langgenius/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-07-22 12:36:14 -04:00
Code Issues 775 Packages Projects Releases 10 Wiki Activity

Question Classifier Output Differs from Flow Process Visualization #11840

New Issue
Closed
opened 2026-02-21 19:04:20 -05:00 by yindo · 1 comment
yindo commented 2026-02-21 19:04:20 -05:00
Owner
Copy Link
Copy Source

Originally created by @1Ckpwee on GitHub (Mar 19, 2025).

Self Checks

  • This is only for bug report, if you would like to ask a question, please head to Discussions.
  • I have searched for existing issues search for existing issues, including closed ones.
  • I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
  • [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）
  • Please do not modify this template :) and fill in all the required fields.

Dify version

1.1.0

Cloud or Self Hosted

Cloud

Steps to reproduce

  1. Create a Chatflow and add a Question Classifier node.
  2. Define three Classes in the classifier (e.g., Class 1, Class 2, Class 3).
  3. Route each Class to a separate output node that prints the respective class name.
  4. Use the Preview feature and enter an input that should trigger Class 1.
  5. Observe the output—it contains Class 1, Class 2, and Class 3 instead of just Class 1.
Image Image Image

✔️ Expected Behavior

The Question Classifier should only trigger the single most relevant class based on the input instead of activating multiple classes simultaneously.

❌ Actual Behavior

it contains Class 1, Class 2, and Class 3 instead of just Class 1.

Originally created by @1Ckpwee on GitHub (Mar 19, 2025). ### Self Checks - [x] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general). - [x] I have searched for existing issues [search for existing issues](https://github.com/langgenius/dify/issues), including closed ones. - [x] I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)). - [x] [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:） - [x] Please do not modify this template :) and fill in all the required fields. ### Dify version 1.1.0 ### Cloud or Self Hosted Cloud ### Steps to reproduce 1. Create a Chatflow and add a Question Classifier node. 2. Define three Classes in the classifier (e.g., Class 1, Class 2, Class 3). 3. Route each Class to a separate output node that prints the respective class name. 4. Use the Preview feature and enter an input that should trigger Class 1. 5. Observe the output—it contains Class 1, Class 2, and Class 3 instead of just Class 1. <img width="672" alt="Image" src="https://github.com/user-attachments/assets/ab6f34c3-6f0b-4714-98fa-be5e7968b648" /> <img width="413" alt="Image" src="https://github.com/user-attachments/assets/cdddc0f6-7650-4062-b70b-8124cde406bf" /> <img width="335" alt="Image" src="https://github.com/user-attachments/assets/35c70ad3-4edc-4ed0-8dc7-92db5017a1bd" /> ### ✔️ Expected Behavior The Question Classifier should only trigger the single most relevant class based on the input instead of activating multiple classes simultaneously. ### ❌ Actual Behavior it contains Class 1, Class 2, and Class 3 instead of just Class 1.
yindo added the 🐞 bug label 2026-02-21 19:04:20 -05:00
yindo closed this issue 2026-02-21 19:04:20 -05:00
yindo commented 2026-02-21 19:04:20 -05:00
Author
Owner
Copy Link
Copy Source

@dosubot[bot] commented on GitHub (Mar 19, 2025):

Hey @1Ckpwee, I'm here to help you with your bug report! I'm a bot that can assist with bugs, answer questions, and even guide you on becoming a contributor. Let's tackle this together! 🤖

I found a similar unsolved discussion that might be relevant to your issue: Question Classification function in workflow maybe abnormal, which was last updated on December 30, 2024 [1].

Regarding your issue, the expected behavior of the Question Classifier node is to output a single class label based on the input, using a large language model to infer the most relevant classification [2]. If the classifier is outputting multiple classes instead of a single one, it could be due to misconfiguration or a bug in the system.

Ensure that the configuration options, such as the inference model and classification labels, are set correctly. The temperature parameter, which influences the randomness of the model's responses, might also affect the classification precision if not set appropriately [3].

If the issue persists, it might be beneficial to check for any recent updates or changes to the Question Classifier node that could impact its behavior, although I couldn't find specific information on recent changes [3].

To continue talking to Dosu, mention @dosu.

Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other

@dosubot[bot] commented on GitHub (Mar 19, 2025): <!-- Greeting --> Hey @1Ckpwee, I'm here to help you with your bug report! I'm a bot that can assist with bugs, answer questions, and even guide you on becoming a contributor. Let's tackle this together! 🤖 <!-- Answer --> I found a similar unsolved discussion that might be relevant to your issue: [Question Classification function in workflow maybe abnormal](https://github.com/langgenius/dify/discussions/8430), which was last updated on December 30, 2024 <sup>[[1]](https://github.com/langgenius/dify/discussions/8430)</sup>. Regarding your issue, the expected behavior of the Question Classifier node is to output a single class label based on the input, using a large language model to infer the most relevant classification <sup>[[2]](https://github.com/langgenius/dify-docs/blob/main/jp/guides/workflow/node/question-classifier.md)</sup>. If the classifier is outputting multiple classes instead of a single one, it could be due to misconfiguration or a bug in the system. Ensure that the configuration options, such as the inference model and classification labels, are set correctly. The `temperature` parameter, which influences the randomness of the model's responses, might also affect the classification precision if not set appropriately <sup>[[3]](https://github.com/langgenius/dify/blob/main/web/app/components/workflow/nodes/question-classifier/default.ts)</sup>. If the issue persists, it might be beneficial to check for any recent updates or changes to the Question Classifier node that could impact its behavior, although I couldn't find specific information on recent changes <sup>[[3]](https://github.com/langgenius/dify/blob/main/web/app/components/workflow/nodes/question-classifier/default.ts)</sup>. <!-- Dosu Comment Footer --> *To continue talking to [Dosu](https://dosu.dev), mention @dosu.* --- <sup>Help Dosu learn! Give it feedback: [Great Response](https://app.dosu.dev/response-feedback/249a12e0-9ddc-4453-b986-ffa43468b05d?feedback_type=great_response) | [Irrelevant Answer](https://app.dosu.dev/response-feedback/249a12e0-9ddc-4453-b986-ffa43468b05d?feedback_type=irrelevant_answer) | [Incorrect Sources](https://app.dosu.dev/response-feedback/249a12e0-9ddc-4453-b986-ffa43468b05d?feedback_type=incorrect_sources) | [Too Verbose](https://app.dosu.dev/response-feedback/249a12e0-9ddc-4453-b986-ffa43468b05d?feedback_type=too_verbose) | [Hallucination](https://app.dosu.dev/response-feedback/249a12e0-9ddc-4453-b986-ffa43468b05d?feedback_type=hallucination) | [Bug Report](https://app.dosu.dev/response-feedback/249a12e0-9ddc-4453-b986-ffa43468b05d?feedback_type=bug_report) | [Other](https://app.dosu.dev/response-feedback/249a12e0-9ddc-4453-b986-ffa43468b05d?feedback_type=other)</sup>
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Knowledge Pipeline
Proposal
WIP
cloud
dependencies
feat:webapp
good first issue
outdated
performance
pull-request
Mirrored from GitHub Pull Request
 python:uv
refactor
size:L
stale
status: accepting prs
status: blocked
status: in discussion
stt & tts
vulnerablity
web
wont-fix
🌊 feat:workflow
🌚 invalid
🐞 bug
👻 feat:rag
💪 enhancement
📚 documentation
🔨 feat:tools
🙋‍♂️ question
🙏 help wanted
🤔 cant-reproduce
🤖 feat:agent
No Label 🐞 bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langgenius/dify#11840
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?