langgenius
The problem classifier will perform on all nodes
Self Checks
- [x] This is only for bug report, if you would like to ask a question, please head to Discussions.
- [x] I have searched for existing issues search for existing 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).
- [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
Self Hosted (Source)
Steps to reproduce
✔️ Expected Behavior
Run only the correct branch according to the green path
❌ Actual Behavior
The problem classifier shows correctly on the flowchart, but it actually runs all branches.
The DSL file
app:
description: ''
icon: 🤖
icon_background: '#FFEAD5'
mode: advanced-chat
name: Minimum test unit
use_icon_as_answer_icon: false
dependencies:
- current_identifier: null
type: marketplace
value:
marketplace_plugin_unique_identifier: langgenius/tongyi:0.0.12@0e4f4caf2002b8fa6de179abf3e8b91b640324921982e9999d96967474f43e18
kind: app
version: 0.1.5
workflow:
conversation_variables: []
environment_variables: []
features:
file_upload:
allowed_file_extensions:
- .JPG
- .JPEG
- .PNG
- .GIF
- .WEBP
- .SVG
allowed_file_types:
- image
allowed_file_upload_methods:
- local_file
- remote_url
enabled: false
fileUploadConfig:
audio_file_size_limit: 50
batch_count_limit: 5
file_size_limit: 15
image_file_size_limit: 10
video_file_size_limit: 100
workflow_file_upload_limit: 10
image:
enabled: false
number_limits: 3
transfer_methods:
- local_file
- remote_url
number_limits: 3
opening_statement: ''
retriever_resource:
enabled: true
sensitive_word_avoidance:
enabled: false
speech_to_text:
enabled: false
suggested_questions: []
suggested_questions_after_answer:
enabled: false
text_to_speech:
enabled: false
language: ''
voice: ''
graph:
edges:
- data:
isInIteration: false
isInLoop: false
sourceType: start
targetType: answer
id: 1742382361944-source-1742382401920-target
source: '1742382361944'
sourceHandle: source
target: '1742382401920'
targetHandle: target
type: custom
zIndex: 0
- data:
isInIteration: false
isInLoop: false
sourceType: answer
targetType: question-classifier
id: 1742382401920-source-1742382406742-target
source: '1742382401920'
sourceHandle: source
target: '1742382406742'
targetHandle: target
type: custom
zIndex: 0
- data:
isInIteration: false
isInLoop: false
sourceType: question-classifier
targetType: llm
id: 1742382406742-1-1742382460532-target
source: '1742382406742'
sourceHandle: '1'
target: '1742382460532'
targetHandle: target
type: custom
zIndex: 0
- data:
isInIteration: false
isInLoop: false
sourceType: question-classifier
targetType: question-classifier
id: 1742382406742-1-1742382480077-target
source: '1742382406742'
sourceHandle: '1'
target: '1742382480077'
targetHandle: target
type: custom
zIndex: 0
- data:
isInIteration: false
isInLoop: false
sourceType: question-classifier
targetType: answer
id: 1742382480077-1-1742382531085-target
source: '1742382480077'
sourceHandle: '1'
target: '1742382531085'
targetHandle: target
type: custom
zIndex: 0
- data:
isInIteration: false
isInLoop: false
sourceType: question-classifier
targetType: answer
id: 1742382480077-2-1742382534798-target
source: '1742382480077'
sourceHandle: '2'
target: '1742382534798'
targetHandle: target
type: custom
zIndex: 0
- data:
isInIteration: false
isInLoop: false
sourceType: question-classifier
targetType: answer
id: 1742382480077-1742382525856-1742382538517-target
source: '1742382480077'
sourceHandle: '1742382525856'
target: '1742382538517'
targetHandle: target
type: custom
zIndex: 0
nodes:
- data:
desc: ''
selected: false
title: 开始
type: start
variables: []
height: 53
id: '1742382361944'
position:
x: 80
y: 282
positionAbsolute:
x: 80
y: 282
sourcePosition: right
targetPosition: left
type: custom
width: 243
- data:
answer: '### 我正在理解您的问题请稍等🧠🧠🧠
'
desc: ''
selected: false
title: 直接回复
type: answer
variables: []
height: 117
id: '1742382401920'
position:
x: 383
y: 282
positionAbsolute:
x: 383
y: 282
sourcePosition: right
targetPosition: left
type: custom
width: 243
- data:
classes:
- id: '1'
name: 是税务相关问题
- id: '2'
name: 不是税务相关问题
desc: ''
instruction: ''
instructions: ''
model:
completion_params:
temperature: 0.7
mode: chat
name: qwen-max-latest
provider: langgenius/tongyi/tongyi
query_variable_selector:
- '1742382361944'
- sys.query
selected: false
title: 问题分类器
topics: []
type: question-classifier
vision:
enabled: false
height: 171
id: '1742382406742'
position:
x: 686
y: 282
positionAbsolute:
x: 686
y: 282
selected: false
sourcePosition: right
targetPosition: left
type: custom
width: 243
- data:
context:
enabled: false
variable_selector: []
desc: ''
model:
completion_params:
temperature: 0.7
mode: chat
name: qwen-max-latest
provider: langgenius/tongyi/tongyi
prompt_template:
- id: 3ccf5c41-e58c-45eb-b610-6d7002d31d24
role: system
text: 你是一个非常有经验的税务专家，解答用户的问题
- id: be521ed7-943d-4500-9503-55c5fa43f803
role: user
text: '{{#sys.query#}}'
selected: false
title: LLM
type: llm
variables: []
vision:
enabled: false
height: 89
id: '1742382460532'
position:
x: 989
y: 282
positionAbsolute:
x: 989
y: 282
selected: true
sourcePosition: right
targetPosition: left
type: custom
width: 243
- data:
classes:
- id: '1'
name: 增值税
- id: '2'
name: 印花税
- id: '1742382525856'
name: 其他
desc: ''
instruction: ''
instructions: ''
model:
completion_params:
temperature: 0.7
mode: chat
name: qwen-max-latest
provider: langgenius/tongyi/tongyi
query_variable_selector:
- '1742382361944'
- sys.query
selected: false
title: 问题分类器 2
topics: []
type: question-classifier
vision:
enabled: false
height: 209
id: '1742382480077'
position:
x: 989
y: 410
positionAbsolute:
x: 989
y: 410
selected: false
sourcePosition: right
targetPosition: left
type: custom
width: 243
- data:
answer: '增值税:{{#sys.query#}}
'
desc: ''
selected: false
title: 直接回复 2
type: answer
variables: []
height: 104
id: '1742382531085'
position:
x: 1292
y: 410
positionAbsolute:
x: 1292
y: 410
selected: false
sourcePosition: right
targetPosition: left
type: custom
width: 243
- data:
answer: '印花税:{{#sys.query#}}
'
desc: ''
selected: false
title: 直接回复 3
type: answer
variables: []
height: 104
id: '1742382534798'
position:
x: 1292
y: 550
positionAbsolute:
x: 1292
y: 550
selected: false
sourcePosition: right
targetPosition: left
type: custom
width: 243
- data:
answer: '其他:{{#sys.query#}}
'
desc: ''
selected: false
title: 直接回复 4
type: answer
variables: []
height: 104
id: '1742382538517'
position:
x: 1292
y: 690
positionAbsolute:
x: 1292
y: 690
selected: false
sourcePosition: right
targetPosition: left
type: custom
width: 243
viewport:
x: 97.80114691770643
y: 50.40037330669551
zoom: 0.757858283255199
Hey @CarltonHere! I'm here to help you with your DSL file issue. Let's squash those bugs together! 🤖
I found a similar issue related to the problem classifier running all branches instead of only the correct one. In Dify 0.13.2, a known issue occurs when multiple paths from an IF/ELSE statement connect to the same problem classifier, causing it to execute all paths. A temporary workaround is to add an empty component (like assigning a value to an empty variable) in all paths between IF/ELSE branches to prevent this behavior [1].
Additionally, there was a closed issue where the chatbot workflow returned answers containing other branches. This was resolved by fixing the answer node stream processing in conditional branches [2]. You might want to check if a similar fix applies to your version or if a workaround can be implemented based on these insights.
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] avatar](https://avatars.githubusercontent.com/in/324583?v=4 "Published by a pub.dev verified publisher"){kind=small}
Hey @CarltonHere! I'm here to help you with your DSL file issue. Let's squash those bugs together! 🤖
I found a similar issue related to the problem classifier running all branches instead of only the correct one. In Dify 0.13.2, a known issue occurs when multiple paths from an IF/ELSE statement connect to the same problem classifier, causing it to execute all paths. A temporary workaround is to add an empty component (like assigning a value to an empty variable) in all paths between IF/ELSE branches to prevent this behavior [1].
Additionally, there was a closed issue where the chatbot workflow returned answers containing other branches. This was resolved by fixing the answer node stream processing in conditional branches [2]. You might want to check if a similar fix applies to your version or if a workaround can be implemented based on these insights.
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
Thanks, adding an empty component does work, but it can be confirmed that this is a BUG, hoping to be thoroughly resolved.
i have the same issue after updated to Cloud PRO v1.1 Also the if/else node always show all answers
For me also works to add an empty variable assigner before EVERY ANSWER inside my workflow... A lot of time wasted to stop everything and manually fix this bug inside all workflows...