mirror of
https://github.com/microsoft/autogen.git
synced 2026-04-20 03:02:16 -04:00
"model" to "chat completion" (#4804)
Co-authored-by: Jack Gerrits <jackgerrits@users.noreply.github.com>
This commit is contained in:
Eric Zhu
@@ -1,137 +1,137 @@
|
||||
{
|
||||
"cells": [
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"# Component config\n",
|
||||
"\n",
|
||||
"AutoGen components are able to be declaratively configured in a generic fashion. This is to support configuration based experiences, such as AutoGen studio, but it is also useful for many other scenarios.\n",
|
||||
"\n",
|
||||
"The system that provides this is called \"component configuration\". In AutoGen, a component is simply something that can be created from a config object and itself can be dumped to a config object. In this way, you can define a component in code and then get the config object from it.\n",
|
||||
"\n",
|
||||
"This system is generic and allows for components defined outside of AutoGen itself (such as extensions) to be configured in the same way.\n",
|
||||
"\n",
|
||||
"## How does this differ from state?\n",
|
||||
"\n",
|
||||
"This is a very important point to clarify. When we talk about serializing an object, we must include *all* data that makes that object itself. Including things like message history etc. When deserializing from serialized state, you must get back the *exact* same object. This is not the case with component configuration.\n",
|
||||
"\n",
|
||||
"Component configuration should be thought of as the blueprint for an object, and can be stamped out many times to create many instances of the same configured object.\n",
|
||||
"\n",
|
||||
"## Usage\n",
|
||||
"\n",
|
||||
"If you have a component in Python and want to get the config for it, simply call {py:meth}`~autogen_core.ComponentConfig.dump_component` on it. The resulting object can be passed back into {py:meth}`~autogen_core.ComponentLoader.load_component` to get the component back.\n",
|
||||
"\n",
|
||||
"### Loading a component from a config\n",
|
||||
"\n",
|
||||
"To load a component from a config object, you can use the {py:meth}`~autogen_core.ComponentLoader.load_component` method. This method will take a config object and return a component object. It is best to call this method on the interface you want. For example to load a model client:\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"from autogen_core.models import ChatCompletionClient\n",
|
||||
"\n",
|
||||
"config = {\n",
|
||||
" \"provider\": \"openai_model_client\",\n",
|
||||
" \"config\": {\"model\": \"gpt-4o\"},\n",
|
||||
"}\n",
|
||||
"\n",
|
||||
"client = ChatCompletionClient.load_component(config)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Creating a component class\n",
|
||||
"\n",
|
||||
"To add component functionality to a given class:\n",
|
||||
"\n",
|
||||
"1. Add a call to {py:meth}`~autogen_core.Component` in the class inheritance list.\n",
|
||||
"2. Implment the {py:meth}`~autogen_core.ComponentConfigImpl._to_config` and {py:meth}`~autogen_core.ComponentConfigImpl._from_config` methods\n",
|
||||
"\n",
|
||||
"For example:"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"from autogen_core import Component\n",
|
||||
"from pydantic import BaseModel\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"class Config(BaseModel):\n",
|
||||
" value: str\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"class MyComponent(Component[Config]):\n",
|
||||
" component_type = \"custom\"\n",
|
||||
" component_config_schema = Config\n",
|
||||
"\n",
|
||||
" def __init__(self, value: str):\n",
|
||||
" self.value = value\n",
|
||||
"\n",
|
||||
" def _to_config(self) -> Config:\n",
|
||||
" return Config(value=self.value)\n",
|
||||
"\n",
|
||||
" @classmethod\n",
|
||||
" def _from_config(cls, config: Config) -> \"MyComponent\":\n",
|
||||
" return cls(value=config.value)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"\n",
|
||||
"## Secrets\n",
|
||||
"\n",
|
||||
"If a field of a config object is a secret value, it should be marked using [`SecretStr`](https://docs.pydantic.dev/latest/api/types/#pydantic.types.SecretStr), this will ensure that the value will not be dumped to the config object.\n",
|
||||
"\n",
|
||||
"For example:\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 1,
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"from pydantic import BaseModel, SecretStr\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"class ClientConfig(BaseModel):\n",
|
||||
" endpoint: str\n",
|
||||
" api_key: SecretStr"
|
||||
]
|
||||
}
|
||||
],
|
||||
"metadata": {
|
||||
"kernelspec": {
|
||||
"display_name": ".venv",
|
||||
"language": "python",
|
||||
"name": "python3"
|
||||
},
|
||||
"language_info": {
|
||||
"codemirror_mode": {
|
||||
"name": "ipython",
|
||||
"version": 3
|
||||
},
|
||||
"file_extension": ".py",
|
||||
"mimetype": "text/x-python",
|
||||
"name": "python",
|
||||
"nbconvert_exporter": "python",
|
||||
"pygments_lexer": "ipython3",
|
||||
"version": "3.12.7"
|
||||
}
|
||||
},
|
||||
"nbformat": 4,
|
||||
"nbformat_minor": 2
|
||||
"cells": [
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"# Component config\n",
|
||||
"\n",
|
||||
"AutoGen components are able to be declaratively configured in a generic fashion. This is to support configuration based experiences, such as AutoGen studio, but it is also useful for many other scenarios.\n",
|
||||
"\n",
|
||||
"The system that provides this is called \"component configuration\". In AutoGen, a component is simply something that can be created from a config object and itself can be dumped to a config object. In this way, you can define a component in code and then get the config object from it.\n",
|
||||
"\n",
|
||||
"This system is generic and allows for components defined outside of AutoGen itself (such as extensions) to be configured in the same way.\n",
|
||||
"\n",
|
||||
"## How does this differ from state?\n",
|
||||
"\n",
|
||||
"This is a very important point to clarify. When we talk about serializing an object, we must include *all* data that makes that object itself. Including things like message history etc. When deserializing from serialized state, you must get back the *exact* same object. This is not the case with component configuration.\n",
|
||||
"\n",
|
||||
"Component configuration should be thought of as the blueprint for an object, and can be stamped out many times to create many instances of the same configured object.\n",
|
||||
"\n",
|
||||
"## Usage\n",
|
||||
"\n",
|
||||
"If you have a component in Python and want to get the config for it, simply call {py:meth}`~autogen_core.ComponentConfig.dump_component` on it. The resulting object can be passed back into {py:meth}`~autogen_core.ComponentLoader.load_component` to get the component back.\n",
|
||||
"\n",
|
||||
"### Loading a component from a config\n",
|
||||
"\n",
|
||||
"To load a component from a config object, you can use the {py:meth}`~autogen_core.ComponentLoader.load_component` method. This method will take a config object and return a component object. It is best to call this method on the interface you want. For example to load a model client:\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"from autogen_core.models import ChatCompletionClient\n",
|
||||
"\n",
|
||||
"config = {\n",
|
||||
" \"provider\": \"openai_chat_completion_client\",\n",
|
||||
" \"config\": {\"model\": \"gpt-4o\"},\n",
|
||||
"}\n",
|
||||
"\n",
|
||||
"client = ChatCompletionClient.load_component(config)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Creating a component class\n",
|
||||
"\n",
|
||||
"To add component functionality to a given class:\n",
|
||||
"\n",
|
||||
"1. Add a call to {py:meth}`~autogen_core.Component` in the class inheritance list.\n",
|
||||
"2. Implment the {py:meth}`~autogen_core.ComponentConfigImpl._to_config` and {py:meth}`~autogen_core.ComponentConfigImpl._from_config` methods\n",
|
||||
"\n",
|
||||
"For example:"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"from autogen_core import Component\n",
|
||||
"from pydantic import BaseModel\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"class Config(BaseModel):\n",
|
||||
" value: str\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"class MyComponent(Component[Config]):\n",
|
||||
" component_type = \"custom\"\n",
|
||||
" component_config_schema = Config\n",
|
||||
"\n",
|
||||
" def __init__(self, value: str):\n",
|
||||
" self.value = value\n",
|
||||
"\n",
|
||||
" def _to_config(self) -> Config:\n",
|
||||
" return Config(value=self.value)\n",
|
||||
"\n",
|
||||
" @classmethod\n",
|
||||
" def _from_config(cls, config: Config) -> \"MyComponent\":\n",
|
||||
" return cls(value=config.value)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"\n",
|
||||
"## Secrets\n",
|
||||
"\n",
|
||||
"If a field of a config object is a secret value, it should be marked using [`SecretStr`](https://docs.pydantic.dev/latest/api/types/#pydantic.types.SecretStr), this will ensure that the value will not be dumped to the config object.\n",
|
||||
"\n",
|
||||
"For example:\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 1,
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"from pydantic import BaseModel, SecretStr\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"class ClientConfig(BaseModel):\n",
|
||||
" endpoint: str\n",
|
||||
" api_key: SecretStr"
|
||||
]
|
||||
}
|
||||
],
|
||||
"metadata": {
|
||||
"kernelspec": {
|
||||
"display_name": ".venv",
|
||||
"language": "python",
|
||||
"name": "python3"
|
||||
},
|
||||
"language_info": {
|
||||
"codemirror_mode": {
|
||||
"name": "ipython",
|
||||
"version": 3
|
||||
},
|
||||
"file_extension": ".py",
|
||||
"mimetype": "text/x-python",
|
||||
"name": "python",
|
||||
"nbconvert_exporter": "python",
|
||||
"pygments_lexer": "ipython3",
|
||||
"version": "3.12.7"
|
||||
}
|
||||
},
|
||||
"nbformat": 4,
|
||||
"nbformat_minor": 2
|
||||
}
|
||||
|
||||
@@ -40,9 +40,9 @@ def _type_to_provider_str(t: type) -> str:
|
||||
|
||||
|
||||
WELL_KNOWN_PROVIDERS = {
|
||||
"azure_openai_model_client": "autogen_ext.models.openai.AzureOpenAIChatCompletionClient",
|
||||
"azure_openai_chat_completion_client": "autogen_ext.models.openai.AzureOpenAIChatCompletionClient",
|
||||
"AzureOpenAIChatCompletionClient": "autogen_ext.models.openai.AzureOpenAIChatCompletionClient",
|
||||
"openai_model_client": "autogen_ext.models.openai.OpenAIChatCompletionClient",
|
||||
"openai_chat_completion_client": "autogen_ext.models.openai.OpenAIChatCompletionClient",
|
||||
"OpenAIChatCompletionClient": "autogen_ext.models.openai.OpenAIChatCompletionClient",
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user