🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org
Python agentai-societiesartificial-intelligencecommunicative-aicooperative-aideep-learninglarge-language-modelsmulti-agent-systemsnatural-language-processing
<a id="camel.data_collector.base"></a>
<a id="camel.data_collector.base.CollectorData"></a>
## CollectorData
```python
class CollectorData:
```
<a id="camel.data_collector.base.CollectorData.__init__"></a>
### __init__
```python
def __init__(
self,
id: UUID,
name: str,
role: Literal['user', 'assistant', 'system', 'tool'],
message: Optional[str] = None,
function_call: Optional[Dict[str, Any]] = None
):
```
Create a data item store information about a message.
Used by the data collector.
**Parameters:**
- **id** (UUID): The id of the message.
- **name** (str): The name of the agent.
- **role** (`Literal["user", "assistant", "system", "function"]`): The role of the message.
- **message** (Optional[str], optional): The message. (default: :obj:`None`)
- **function_call** (Optional[Dict[str, Any]], optional): The function call. (default: :obj:`None`)
<a id="camel.data_collector.base.CollectorData.from_context"></a>
### from_context
```python
def from_context(name, context: Dict[str, Any]):
```
Create a data collector from a context.
**Parameters:**
- **name** (str): The name of the agent.
- **context** (Dict[str, Any]): The context.
**Returns:**
CollectorData: The data collector.
<a id="camel.data_collector.base.BaseDataCollector"></a>
## BaseDataCollector
```python
class BaseDataCollector(ABC):
```
Base class for data collectors.
<a id="camel.data_collector.base.BaseDataCollector.__init__"></a>
### __init__
```python
def __init__(self):
```
Create a data collector.
<a id="camel.data_collector.base.BaseDataCollector.step"></a>
### step
```python
def step(
self,
role: Literal['user', 'assistant', 'system', 'tool'],
name: Optional[str] = None,
message: Optional[str] = None,
function_call: Optional[Dict[str, Any]] = None
):
```
Record a message.
**Parameters:**
- **role** (`Literal["user", "assistant", "system", "tool"]`): The role of the message.
- **name** (Optional[str], optional): The name of the agent. (default: :obj:`None`)
- **message** (Optional[str], optional): The message to record. (default: :obj:`None`)
- **function_call** (Optional[Dict[str, Any]], optional): The function call to record. (default: :obj:`None`)
**Returns:**
Self: The data collector.
<a id="camel.data_collector.base.BaseDataCollector.record"></a>
### record
```python
def record(self, agent: Union[List[ChatAgent], ChatAgent]):
```
Record agents.
**Parameters:**
- **agent** (Union[List[ChatAgent], ChatAgent]): The agent(s) to inject.
<a id="camel.data_collector.base.BaseDataCollector.start"></a>
### start
```python
def start(self):
```
Start recording.
<a id="camel.data_collector.base.BaseDataCollector.stop"></a>
### stop
```python
def stop(self):
```
Stop recording.
<a id="camel.data_collector.base.BaseDataCollector.recording"></a>
### recording
```python
def recording(self):
```
Whether the collector is recording.
<a id="camel.data_collector.base.BaseDataCollector.reset"></a>
### reset
```python
def reset(self, reset_agents: bool = True):
```
Reset the collector.
**Parameters:**
- **reset_agents** (bool, optional): Whether to reset the agents. Defaults to True.
<a id="camel.data_collector.base.BaseDataCollector.convert"></a>
### convert
```python
def convert(self):
```
Convert the collected data.
<a id="camel.data_collector.base.BaseDataCollector.llm_convert"></a>
### llm_convert
```python
def llm_convert(self, converter: Any, prompt: Optional[str] = None):
```
Convert the collected data.
<a id="camel.data_collector.base.BaseDataCollector.get_agent_history"></a>
### get_agent_history
```python
def get_agent_history(self, name: str):
```
Get the message history of an agent.
**Parameters:**
- **name** (str): The name of the agent.
**Returns:**
List[CollectorData]: The message history of the agent