github/autogen
1
1
Fork 0
mirror of https://github.com/microsoft/autogen.git synced 2026-04-20 03:02:16 -04:00
Code Issues Packages Projects Releases Wiki Activity

support async in agents (#1178)

Browse Source 
* Make auto reply method pluggable

* support async

* async

* allow richer trigger types

* test list

* rename key
This commit is contained in:
Chi Wang
2023-08-07 18:34:47 -07:00
committed by GitHub
parent a603e6dddc
commit c44d2f4a01
15 changed files with 1111 additions and 39 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
website/docs/Use-Cases/Auto-Generation.md
View File

@@ -11,11 +11,13 @@
The package is under active development with more features upcoming.
## Agents (Experimental)
## Agents
[`flaml.autogen.agentchat`](/docs/reference/autogen/agentchat/agent) contains an experimental implementation of interactive agents which can adapt to human or simulated feedback. This subpackage is under active development.
[`flaml.autogen.agentchat`](/docs/reference/autogen/agentchat/agent) offers conversable agents which can adapt to human or simulated feedback. This subpackage is under active development.
We have designed a generic `ResponsiveAgent` class for Agents that are capable of communicating with each other through the exchange of messages to collaboratively finish a task. An agent can communicate with other agents and perform actions. Different agents can differ in what actions they perform after receiving messages. Two representative subclasses are `AssistantAgent` and `UserProxyAgent`.
### Basic Concept
We have designed a generic `ResponsiveAgent` class for Agents that are capable of conversing with each other through the exchange of messages to collaboratively finish a task. An agent can communicate with other agents and perform actions. Different agents can differ in what actions they perform after receiving messages. Two representative subclasses are `AssistantAgent` and `UserProxyAgent`.
- `AssistantAgent`. Designed to act as an assistant by responding to user requests. It could write Python code (in a Python coding block) for a user to execute when a message (typically a description of a task that needs to be solved) is received. Under the hood, the Python code is written by LLM (e.g., GPT-4). It can also receive the execution results and suggest code with bug fix. Its behavior can be altered by passing a new system message. The LLM [inference](#enhanced-inference) configuration can be configured via `llm_config`.
- `UserProxyAgent`. Serves as a proxy for the human user. Upon receiving a message, the UserProxyAgent will either solicit the human user's input or prepare an automatically generated reply. The chosen action depends on the settings of the `human_input_mode` and `max_consecutive_auto_reply` when the `UserProxyAgent` instance is constructed, and whether a human user input is available.
@@ -25,6 +27,8 @@ When `llm_config` is set to a dict, `UserProxyAgent` can generate replies using
The auto-reply capability of `ResponsiveAgent` allows for more autonomous multi-agent communication while retaining the possibility of human intervention.
One can also easily extend it by registering auto_reply functions with the `register_auto_reply()` method.
### Basic Example
Example usage of the agents to solve a task with code:
```python
from flaml.autogen import AssistantAgent, UserProxyAgent
@@ -53,13 +57,14 @@ In the example above, we create an AssistantAgent named "assistant" to serve as
Please find a visual illustration of how UserProxyAgent and AssistantAgent collaboratively solve the above task below:
![Agent Chat Example](images/agent_example.png)
#### Human Input Mode
### Human Input Mode
The `human_input_mode` parameter of `UserProxyAgent` controls the behavior of the agent when it receives a message. It can be set to `"NEVER"`, `"ALWAYS"`, or `"TERMINATE"`.
- Under the mode `human_input_mode="NEVER"`, the multi-turn conversation between the assistant and the user_proxy stops when the number of auto-reply reaches the upper limit specified by `max_consecutive_auto_reply` or the received message is a termination message according to `is_termination_msg`.
- When `human_input_mode` is set to `"ALWAYS"`, the user proxy agent solicits human input every time a message is received; and the conversation stops when the human input is "exit", or when the received message is a termination message and no human input is provided.
- When `human_input_mode` is set to `"TERMINATE"`, the user proxy agent solicits human input only when a termination message is received or the number of auto replies reaches `max_consecutive_auto_reply`.
#### Function Calling
### Function Calling
To leverage [function calling capability of OpenAI's Chat Completions API](https://openai.com/blog/function-calling-and-other-api-updates?ref=upstract.com), one can pass in a list of callable functions or class methods to `UserProxyAgent`, which corresponds to the description of functions passed to OpenAI's API.
Example usage of the agents to solve a task with function calling feature:
@@ -141,6 +146,8 @@ user_proxy.initiate_chat(
)
```
### Notebook Examples
*Interested in trying it yourself? Please check the following notebook examples:*
* [Automated Task Solving with Code Generation, Execution & Debugging](https://github.com/microsoft/FLAML/blob/main/notebook/autogen_agentchat_auto_feedback_from_code_execution.ipynb)
@@ -158,6 +165,8 @@ user_proxy.initiate_chat(
* [Automated Task Solving by Group Chat](https://github.com/microsoft/FLAML/blob/main/notebook/autogen_agentchat_groupchat.ipynb)
* [Automated Continual Learning from New Data](https://github.com/microsoft/FLAML/blob/main/notebook/autogen_agentchat_stream.ipynb)
## Enhanced Inference
One can use [`flaml.autogen.Completion.create`](/docs/reference/autogen/oai/completion#create) to perform inference.
@@ -546,6 +555,7 @@ The compact history is more efficient and the individual API call history contai
[`flaml.autogen.math_utils`](/docs/reference/autogen/math_utils) offers utilities for math problems, such as:
- a [eval_math_responses](/docs/reference/autogen/math_utils#eval_math_responses) function to select a response using voting, and check if the final answer is correct if the canonical solution is provided.
## For Further Reading
*Interested in the research that leads to this package? Please check the following papers.*
* [Cost-Effective Hyperparameter Optimization for Large Language Model Generation Inference](https://arxiv.org/abs/2303.04673). Chi Wang, Susan Xueqing Liu, Ahmed H. Awadallah. ArXiv preprint arXiv:2303.04673 (2023).