Comments (4)
@tuhahaha Let's discuss how we can address this issue when you return from your vacation.
A little more background on the cause of this issue:
When a user clicks the Add button, (1) the URL is saved to a temporary file, and (2) the chatbot will look at that temporary file. The current implementation assumes that Step 1 finishes before Step 2 starts. However, this doesn't always seem to be the case. For example, what if Step 1 is blocked by some other operation? What if the write operation fails? In such scenarios, Step 2 will fail with an
IndexError: list index out of range
.
from qwen-agent.
I suspect that the frequent cause of this issue is either one of the following:
- (More likely) The user is clicking the Add button while having the popup page open. This can lead to a race condition. We need to add some precautions to prevent this.
- (Less likely) We are having a permission issue when dealing with that temporary file. Perhaps even the directory is not created properly.
from qwen-agent.
Oh, this is a different issue, not the same one that I am talking about in my previous comment. The IndexError occurs when executing te following code:
...
data['content'] = parse_html_bs(...) # Using bs4 inside to parse html content received from the chrome extension.
...
extract = data['content'][0]['metadata']['title'] # Here!
@tuhahaha It looks like bs4 fails to retrieve the content of the page here. I'm still trying to reproduce this issue and figure out why it occurs.
@tuhahaha Btw, we need to automatically detect the language and use English prompts when dealing with English requests.
from qwen-agent.
I just pushed a quick fix. Please pull the latest code and see if it works for you.
Details:
- I managed to reproduce the issue on a Windows machine. WSL2 works fine, but native Windows does not.
- On my machine, this is related to the str codec, which is a mess on Windows since it does not always use UTF-8 and may instead use other encodings such as GBK or Latin-1.
from qwen-agent.
Related Issues (20)
- 请问agent 能否支持类似langchain中显示intermediate step/scratchpad plan等功能?
- 私有云部署了chatglm3-6B模型,通过request.post(data)调用模型输出。这种调用方式如何适配到qwen-agent里? HOT 2
- 关于system_message和PROMPT_REACT的问题 HOT 1
- 提高Assistant知识检索速度的方法? HOT 1
- 您好,Mr,Ma。我想把代码转成非流式输出,模型输出了最终回复后报错,我看了下代码这里,拿到模型的output是如下形式,而代码里判断的却是是否是list形式。是不是对于非流式,对模型的输出还要做一定处理? HOT 5
- 您好,咨询一下,agent内部调用llm的_call_llm(...)是默认非流式调用,我要所有agent都换成非流式调用,有相关参考可以查看吗?需要修改Agent里的返回类型 HOT 1
- Move tools like `code_interpreter` in an optional package
- 确认一下function call微调所使用的chat_template
- 如何写prompt让ReActChat在调用code interpreter的时候,按照指定的步骤查询excel表格?
- 运行example报错 HOT 5
- 有什么措施可以使模型更稳定的调用插件 HOT 1
- [报错] 添加文件后,进行对话时报错 HOT 1
- 能支持langserve吗?
- 如何改写可以是的Agent可以非流式调用LLM,更改了后很多数据类型有错误。 HOT 1
- 关于sse返回结构, 这两种流式返回方式有区别吗? HOT 1
- ParallelDocQA有没有做概要总结的能力? HOT 2
- 不支持function calling功能?还是姿势不对? HOT 5
- TypeError: typing.ClassVar[pydantic.config.ConfigDict] is not valid as type argument
- ParellelDocQA的文档问答能力
- ParallelDocQA可以限制相关句子的长度或数量吗? HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from qwen-agent.