Skip to content

Commit

Permalink
Update RetrieveChat blog to fix some outdated APIs (microsoft#2273)
Browse files Browse the repository at this point in the history 
* Fix rag blog

* Update wording
  • Loading branch information
@thinkall
thinkall authored Apr 4, 2024
1 parent 4d688a5 commit 46bee1f
Show file tree
Hide file tree
Showing 2 changed files with 59 additions and 65 deletions.
29 changes: 12 additions & 17 deletions notebook/agentchat_groupchat_RAG.ipynb
View file
Original file line number Diff line number Diff line change
Expand Up @@ -79,13 +79,6 @@
"metadata": {},
"outputs": [],
"source": [
"llm_config = {\n",
" \"timeout\": 60,\n",
" \"temperature\": 0,\n",
" \"config_list\": config_list,\n",
"}\n",
"\n",
"\n",
"def termination_msg(x):\n",
" return isinstance(x, dict) and \"TERMINATE\" == str(x.get(\"content\", \"\"))[-9:].upper()\n",
"\n",
Expand Down Expand Up @@ -117,31 +110,27 @@
" description=\"Assistant who has extra content retrieval power for solving difficult problems.\",\n",
")\n",
"\n",
"\n",
"coder_llm_config = llm_config.copy()\n",
"coder = AssistantAgent(\n",
" name=\"Senior_Python_Engineer\",\n",
" is_termination_msg=termination_msg,\n",
" system_message=\"You are a senior python engineer, you provide python code to answer questions. Reply `TERMINATE` in the end when everything is done.\",\n",
" llm_config={\"config_list\": config_list},\n",
" llm_config={\"config_list\": config_list, \"timeout\": 60, \"temperature\": 0},\n",
" description=\"Senior Python Engineer who can write code to solve problems and answer questions.\",\n",
")\n",
"\n",
"pm_llm_config = llm_config.copy()\n",
"pm = autogen.AssistantAgent(\n",
" name=\"Product_Manager\",\n",
" is_termination_msg=termination_msg,\n",
" system_message=\"You are a product manager. Reply `TERMINATE` in the end when everything is done.\",\n",
" llm_config={\"config_list\": config_list},\n",
" llm_config={\"config_list\": config_list, \"timeout\": 60, \"temperature\": 0},\n",
" description=\"Product Manager who can design and plan the project.\",\n",
")\n",
"\n",
"reviewer_llm_config = llm_config.copy()\n",
"reviewer = autogen.AssistantAgent(\n",
" name=\"Code_Reviewer\",\n",
" is_termination_msg=termination_msg,\n",
" system_message=\"You are a code reviewer. Reply `TERMINATE` in the end when everything is done.\",\n",
" llm_config={\"config_list\": config_list},\n",
" llm_config={\"config_list\": config_list, \"timeout\": 60, \"temperature\": 0},\n",
" description=\"Code Reviewer who can review the code.\",\n",
")\n",
"\n",
Expand All @@ -161,7 +150,9 @@
" groupchat = autogen.GroupChat(\n",
" agents=[boss_aid, pm, coder, reviewer], messages=[], max_round=12, speaker_selection_method=\"round_robin\"\n",
" )\n",
" manager = autogen.GroupChatManager(groupchat=groupchat, llm_config=llm_config)\n",
" manager = autogen.GroupChatManager(\n",
" groupchat=groupchat, llm_config={\"config_list\": config_list, \"timeout\": 60, \"temperature\": 0}\n",
" )\n",
"\n",
" # Start chatting with boss_aid as this is the user proxy agent.\n",
" boss_aid.initiate_chat(\n",
Expand All @@ -181,7 +172,9 @@
" speaker_selection_method=\"auto\",\n",
" allow_repeat_speaker=False,\n",
" )\n",
" manager = autogen.GroupChatManager(groupchat=groupchat, llm_config=llm_config)\n",
" manager = autogen.GroupChatManager(\n",
" groupchat=groupchat, llm_config={\"config_list\": config_list, \"timeout\": 60, \"temperature\": 0}\n",
" )\n",
"\n",
" # Start chatting with the boss as this is the user proxy agent.\n",
" boss.initiate_chat(\n",
Expand Down Expand Up @@ -233,7 +226,9 @@
" allow_repeat_speaker=False,\n",
" )\n",
"\n",
" manager = autogen.GroupChatManager(groupchat=groupchat, llm_config=llm_config)\n",
" manager = autogen.GroupChatManager(\n",
" groupchat=groupchat, llm_config={\"config_list\": config_list, \"timeout\": 60, \"temperature\": 0}\n",
" )\n",
"\n",
" # Start chatting with the boss as this is the user proxy agent.\n",
" boss.initiate_chat(\n",
Expand Down
95 changes: 47 additions & 48 deletions website/blog/2023-10-18-RetrieveChat/index.mdx
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,8 @@ authors: thinkall
tags: [LLM, RAG]
---

*Last update: April 4, 2024; AutoGen version: v0.2.21*

![RAG Architecture](img/retrievechat-arch.png)

**TL;DR:**
Expand Down Expand Up @@ -57,10 +59,17 @@ pip install "pyautogen[retrievechat]"
RetrieveChat can handle various types of documents. By default, it can process
plain text and PDF files, including formats such as 'txt', 'json', 'csv', 'tsv',
'md', 'html', 'htm', 'rtf', 'rst', 'jsonl', 'log', 'xml', 'yaml', 'yml' and 'pdf'.
If you install [unstructured](https://unstructured-io.github.io/unstructured/installation/full_installation.html)
(`pip install "unstructured[all-docs]"`), additional document types such as 'docx',
If you install [unstructured](https://unstructured-io.github.io/unstructured/installation/full_installation.html),
additional document types such as 'docx',
'doc', 'odt', 'pptx', 'ppt', 'xlsx', 'eml', 'msg', 'epub' will also be supported.

- Install `unstructured` in ubuntu
```bash
sudo apt-get update
sudo apt-get install -y tesseract-ocr poppler-utils
pip install unstructured[all-docs]
```

You can find a list of all supported document types by using `autogen.retrieve_utils.TEXT_FORMATS`.

1. Import Agents
Expand Down Expand Up @@ -90,7 +99,7 @@ ragproxyagent = RetrieveUserProxyAgent(
3. Initialize Chat and ask a question
```python
assistant.reset()
ragproxyagent.initiate_chat(assistant, problem="What is autogen?")
ragproxyagent.initiate_chat(assistant, message=ragproxyagent.message_generator, problem="What is autogen?")
```

Output is like:
Expand Down Expand Up @@ -283,28 +292,6 @@ However, you may want to initialize the chat with another agent in some cases. T
you'll need to call it from a function.

```python
llm_config = {
"functions": [
{
"name": "retrieve_content",
"description": "retrieve content for code generation and question answering.",
"parameters": {
"type": "object",
"properties": {
"message": {
"type": "string",
"description": "Refined message which keeps the original meaning and can be used to retrieve content for code generation and question answering.",
}
},
"required": ["message"],
},
},
],
"config_list": config_list,
"timeout": 60,
"seed": 42,
}

boss = autogen.UserProxyAgent(
name="Boss",
is_termination_msg=termination_msg,
Expand All @@ -328,46 +315,58 @@ coder = AssistantAgent(
name="Senior_Python_Engineer",
is_termination_msg=termination_msg,
system_message="You are a senior python engineer. Reply `TERMINATE` in the end when everything is done.",
llm_config=llm_config,
llm_config={"config_list": config_list, "timeout": 60, "temperature": 0},
)

pm = autogen.AssistantAgent(
name="Product_Manager",
is_termination_msg=termination_msg,
system_message="You are a product manager. Reply `TERMINATE` in the end when everything is done.",
llm_config=llm_config,
llm_config={"config_list": config_list, "timeout": 60, "temperature": 0},
)

reviewer = autogen.AssistantAgent(
name="Code_Reviewer",
is_termination_msg=termination_msg,
system_message="You are a code reviewer. Reply `TERMINATE` in the end when everything is done.",
llm_config=llm_config,
llm_config={"config_list": config_list, "timeout": 60, "temperature": 0},
)

def retrieve_content(message, n_results=3):
boss_aid.n_results = n_results # Set the number of results to be retrieved.
# Check if we need to update the context.
update_context_case1, update_context_case2 = boss_aid._check_update_context(message)
if (update_context_case1 or update_context_case2) and boss_aid.update_context:
boss_aid.problem = message if not hasattr(boss_aid, "problem") else boss_aid.problem
_, ret_msg = boss_aid._generate_retrieve_user_reply(message)
else:
_context = {"problem": message, "n_results": n_results}
ret_msg = boss_aid.message_generator(boss_aid, None, _context)
return ret_msg if ret_msg else message

for agent in [boss, coder, pm, reviewer]:
# register functions for all agents.
agent.register_function(
function_map={
"retrieve_content": retrieve_content,
}
)
def retrieve_content(
message: Annotated[
str,
"Refined message which keeps the original meaning and can be used to retrieve content for code generation and question answering.",
],
n_results: Annotated[int, "number of results"] = 3,
) -> str:
boss_aid.n_results = n_results # Set the number of results to be retrieved.
# Check if we need to update the context.
update_context_case1, update_context_case2 = boss_aid._check_update_context(message)
if (update_context_case1 or update_context_case2) and boss_aid.update_context:
boss_aid.problem = message if not hasattr(boss_aid, "problem") else boss_aid.problem
_, ret_msg = boss_aid._generate_retrieve_user_reply(message)
else:
_context = {"problem": message, "n_results": n_results}
ret_msg = boss_aid.message_generator(boss_aid, None, _context)
return ret_msg if ret_msg else message

for caller in [pm, coder, reviewer]:
d_retrieve_content = caller.register_for_llm(
description="retrieve content for code generation and question answering.", api_style="function"
)(retrieve_content)

for executor in [boss, pm]:
executor.register_for_execution()(d_retrieve_content)

groupchat = autogen.GroupChat(
agents=[boss, coder, pm, reviewer], messages=[], max_round=12
agents=[boss, pm, coder, reviewer],
messages=[],
max_round=12,
speaker_selection_method="round_robin",
allow_repeat_speaker=False,
)

llm_config = {"config_list": config_list, "timeout": 60, "temperature": 0}
manager = autogen.GroupChatManager(groupchat=groupchat, llm_config=llm_config)

# Start chatting with the boss as this is the user proxy agent.
Expand Down

0 comments on commit 46bee1f

Please sign in to comment.