mirror of
https://github.com/microsoft/autogen.git
synced 2025-07-19 15:01:52 +00:00
998840f7e0
It clarifies the missing dependencies of all README.md in python/samples/ - Added explicit mention of required dependencies - Improved instructions for initial setup <!-- Thank you for your contribution! Please review https://microsoft.github.io/autogen/docs/Contribute before opening a pull request. --> <!-- Please add a reviewer to the assignee section when you create a PR. If you don't have the access to it, we will shortly find a reviewer and assign them to your PR. --> ## Why are these changes needed? According to issue #6076, several dependencies were missing from the requirements.txt and not mentioned in the README.md instructions. This change adds the missing installation instructions to ensure that users can run the demo smoothly. ## Related issue number Closes #6076 <!-- For example: "Closes #1234" --> ## Checks - [x] I've included any doc changes needed for <https://microsoft.github.io/autogen/>. See <https://github.com/microsoft/autogen/blob/main/CONTRIBUTING.md> to build and test documentation locally. - [x] I've added tests (if relevant) corresponding to the changes introduced in this PR. - [x] I've made sure all auto checks have passed. Co-authored-by: Eric Zhu <ekzhu@users.noreply.github.com>
71 lines
2.4 KiB
Markdown
71 lines
2.4 KiB
Markdown
# AgentChat App with FastAPI
|
|
|
|
This sample demonstrates how to create a simple chat application using
|
|
[AgentChat](https://microsoft.github.io/autogen/stable/user-guide/agentchat-user-guide/index.html)
|
|
and [FastAPI](https://fastapi.tiangolo.com/).
|
|
|
|
You will be using the following features of AgentChat:
|
|
|
|
1. Agent:
|
|
- `AssistantAgent`
|
|
- `UserProxyAgent` with a custom websocket input function
|
|
2. Team: `RoundRobinGroupChat`
|
|
3. State persistence: `save_state` and `load_state` methods of both agent and team.
|
|
|
|
## Setup
|
|
|
|
Install the required packages with OpenAI support:
|
|
|
|
```bash
|
|
pip install -U "autogen-agentchat" "autogen-ext[openai]" "fastapi" "uvicorn[standard]" "PyYAML"
|
|
```
|
|
|
|
To use models other than OpenAI, see the [Models](https://microsoft.github.io/autogen/stable/user-guide/agentchat-user-guide/tutorial/models.html) documentation.
|
|
|
|
Create a new file named `model_config.yaml` in the same directory as this README file to configure your model settings.
|
|
See `model_config_template.yaml` for an example.
|
|
|
|
## Chat with a single agent
|
|
|
|
To start the FastAPI server for single-agent chat, run:
|
|
|
|
```bash
|
|
python app_agent.py
|
|
```
|
|
|
|
Visit http://localhost:8001 in your browser to start chatting.
|
|
|
|
## Chat with a team of agents
|
|
|
|
To start the FastAPI server for team chat, run:
|
|
|
|
```bash
|
|
python app_team.py
|
|
```
|
|
|
|
Visit http://localhost:8002 in your browser to start chatting.
|
|
|
|
The team also includes a `UserProxyAgent` agent with a custom websocket input function
|
|
that allows the user to send messages to the team from the browser.
|
|
|
|
The team follows a round-robin strategy so each agent will take turns to respond.
|
|
When it is the user's turn, the input box will be enabled.
|
|
Once the user sends a message, the input box will be disabled and the agents
|
|
will take turns to respond.
|
|
|
|
## State persistence
|
|
|
|
The agents and team use the `load_state` and `save_state` methods to load and save
|
|
their state from and to files on each turn.
|
|
For the agent, the state is saved to and loaded from `agent_state.json`.
|
|
For the team, the state is saved to and loaded from `team_state.json`.
|
|
You can inspect the state files to see the state of the agents and team
|
|
once you have chatted with them.
|
|
|
|
When the server restarts, the agents and team will load their state from the state files
|
|
to maintain their state across restarts.
|
|
|
|
Additionally, the apps uses separate JSON files,
|
|
`agent_history.json` and `team_history.json`, to store the conversation history
|
|
for display in the browser.
|