Przewodnik migracji Planner krok po kroku

Ten przewodnik migracji pokazuje, jak przejść z FunctionCallingStepwisePlanner na nowe zalecane podejście do funkcji planowania — Auto Function Calling. Nowe podejście daje wyniki bardziej niezawodnie i używa mniejszej liczby tokenów w porównaniu z FunctionCallingStepwisePlanner.

Generowanie planu

Poniższy kod pokazuje, jak wygenerować nowy plan za pomocą funkcji automatycznego wywoływania przy użyciu polecenia FunctionChoiceBehavior = FunctionChoiceBehavior.Auto(). Po wysłaniu żądania do modelu sztucznej inteligencji plan będzie znajdować się w obiekcie, w ChatHistory którym komunikat z Assistant rolą będzie zawierać listę funkcji (kroków) do wywołania.

Stare podejście:

Kernel kernel = Kernel
    .CreateBuilder()
    .AddOpenAIChatCompletion("gpt-4", Environment.GetEnvironmentVariable("OpenAI__ApiKey"))
    .Build();

FunctionCallingStepwisePlanner planner = new();

FunctionCallingStepwisePlannerResult result = await planner.ExecuteAsync(kernel, "Check current UTC time and return current weather in Boston city.");

ChatHistory generatedPlan = result.ChatHistory;

Nowe podejście:

Kernel kernel = Kernel
    .CreateBuilder()
    .AddOpenAIChatCompletion("gpt-4", Environment.GetEnvironmentVariable("OpenAI__ApiKey"))
    .Build();

IChatCompletionService chatCompletionService = kernel.GetRequiredService<IChatCompletionService>();

ChatHistory chatHistory = [];
chatHistory.AddUserMessage("Check current UTC time and return current weather in Boston city.");

OpenAIPromptExecutionSettings executionSettings = new() { FunctionChoiceBehavior = FunctionChoiceBehavior.Auto() };

await chatCompletionService.GetChatMessageContentAsync(chatHistory, executionSettings, kernel);

ChatHistory generatedPlan = chatHistory;

Wykonanie nowego planu

Poniższy kod pokazuje, jak wykonać nowy plan za pomocą funkcji automatycznego wywoływania przy użyciu polecenia FunctionChoiceBehavior = FunctionChoiceBehavior.Auto(). Takie podejście jest przydatne, gdy potrzebny jest tylko wynik, bez etapów planu. W takim przypadku obiekt Kernel może być użyty do przekazania celu do metody InvokePromptAsync. Wynik wykonania planu będzie znajdować się w FunctionResult obiekcie.

Stare podejście:

Kernel kernel = Kernel
    .CreateBuilder()
    .AddOpenAIChatCompletion("gpt-4", Environment.GetEnvironmentVariable("OpenAI__ApiKey"))
    .Build();

FunctionCallingStepwisePlanner planner = new();

FunctionCallingStepwisePlannerResult result = await planner.ExecuteAsync(kernel, "Check current UTC time and return current weather in Boston city.");

string planResult = result.FinalAnswer;

Nowe podejście:

Kernel kernel = Kernel
    .CreateBuilder()
    .AddOpenAIChatCompletion("gpt-4", Environment.GetEnvironmentVariable("OpenAI__ApiKey"))
    .Build();

OpenAIPromptExecutionSettings executionSettings = new() { FunctionChoiceBehavior = FunctionChoiceBehavior.Auto() };

FunctionResult result = await kernel.InvokePromptAsync("Check current UTC time and return current weather in Boston city.", new(executionSettings));

string planResult = result.ToString();

Wykonanie istniejącego planu

Poniższy kod pokazuje, jak wykonać istniejący plan za pomocą funkcji automatycznego wywoływania przy użyciu polecenia FunctionChoiceBehavior = FunctionChoiceBehavior.Auto(). Takie podejście jest przydatne, gdy ChatHistory jest już obecne (np. przechowywane w pamięci podręcznej) i powinno zostać ponownie wykonane, a końcowy wynik powinien zostać dostarczony przez model sztucznej inteligencji.

Stare podejście:

Kernel kernel = Kernel
    .CreateBuilder()
    .AddOpenAIChatCompletion("gpt-4", Environment.GetEnvironmentVariable("OpenAI__ApiKey"))
    .Build();

FunctionCallingStepwisePlanner planner = new();
ChatHistory existingPlan = GetExistingPlan(); // plan can be stored in database  or cache for reusability.

FunctionCallingStepwisePlannerResult result = await planner.ExecuteAsync(kernel, "Check current UTC time and return current weather in Boston city.", existingPlan);

string planResult = result.FinalAnswer;

Nowe podejście:

Kernel kernel = Kernel
    .CreateBuilder()
    .AddOpenAIChatCompletion("gpt-4", Environment.GetEnvironmentVariable("OpenAI__ApiKey"))
    .Build();

IChatCompletionService chatCompletionService = kernel.GetRequiredService<IChatCompletionService>();

ChatHistory existingPlan = GetExistingPlan(); // plan can be stored in database or cache for reusability.

OpenAIPromptExecutionSettings executionSettings = new() { FunctionChoiceBehavior = FunctionChoiceBehavior.Auto() };

ChatMessageContent result = await chatCompletionService.GetChatMessageContentAsync(existingPlan, executionSettings, kernel);

string planResult = result.Content;

Poniższy kod pokazuje, jak wygenerować nowy plan za pomocą funkcji automatycznego wywoływania przy użyciu polecenia function_choice_behavior = FunctionChoiceBehavior.Auto(). Po wysłaniu żądania do modelu sztucznej inteligencji plan będzie znajdować się w obiekcie, w ChatHistory którym komunikat z Assistant rolą będzie zawierać listę funkcji (kroków) do wywołania.

Stare podejście:

from semantic_kernel import Kernel
from semantic_kernel.connectors.ai.open_ai import AzureChatCompletion
from semantic_kernel.planners.function_calling_stepwise_planner import (
    FunctionCallingStepwisePlanner, 
    FunctionCallingStepwisePlannerResult,
)

kernel = Kernel()
kernel.add_service(AzureChatCompletion())

# Add any plugins to the kernel that the planner will leverage
kernel.add_plugins(...)

planner = FunctionCallingStepwisePlanner(service_id="service_id")

result: FunctionCallingStepwisePlannerResult = await planner.invoke(
    kernel=kernel, 
    question="Check current UTC time and return current weather in Boston city.",
)

generated_plan = result.chat_history

Nowe podejście:

from semantic_kernel import Kernel
from semantic_kernel.connectors.ai import FunctionChoiceBehavior
from semantic_kernel.connectors.ai.open_ai import AzureChatCompletion, AzureChatPromptExecutionSettings
from semantic_kernel.contents import ChatHistory

chat_completion_service = AzureChatCompletion()

chat_history = ChatHistory()
chat_hitory.add_user_message("Check current UTC time and return current weather in Boston city.")

request_settings = AzureChatPromptExecutionSettings(function_choice_behavior=FunctionChoiceBehavior.Auto())

# Add any plugins to the kernel that the planner will leverage
kernel = Kernel()
kernel.add_plugins(...)

response = await chat_completion_service.get_chat_message_content(
    chat_history=chat_history,
    settings=request_settings,
    kernel=kernel,
)
print(response)

# The generated plan is now contained inside of `chat_history`.

Wykonanie nowego planu

Poniższy kod pokazuje, jak wykonać nowy plan za pomocą funkcji automatycznego wywoływania przy użyciu polecenia function_choice_behavior = FunctionChoiceBehavior.Auto(). Takie podejście jest przydatne, gdy tylko wynik jest potrzebny bez kroków planu. W takim przypadku obiekt Kernel może służyć do przekazania celu do metody invoke_prompt. Wynik wykonania planu będzie znajdować się w FunctionResult obiekcie.

Stare podejście:

from semantic_kernel import Kernel
from semantic_kernel.connectors.ai.open_ai import AzureChatCompletion
from semantic_kernel.planners.function_calling_stepwise_planner import (
    FunctionCallingStepwisePlanner, 
    FunctionCallingStepwisePlannerResult,
)

kernel = Kernel()
kernel.add_service(AzureChatCompletion())

# Add any plugins to the kernel that the planner will leverage
kernel.add_plugins(...)

planner = FunctionCallingStepwisePlanner(service_id="service_id")

result: FunctionCallingStepwisePlannerResult = await planner.invoke(
    kernel=kernel, 
    question="Check current UTC time and return current weather in Boston city.",
)

print(result.final_answer)

Nowe podejście:

from semantic_kernel import Kernel
from semantic_kernel.connectors.ai import FunctionChoiceBehavior
from semantic_kernel.connectors.ai.open_ai import AzureChatCompletion, AzureChatPromptExecutionSettings
from semantic_kernel.contents import ChatHistory
from semantic_kernel.functions import KernelArguments

kernel = Kernel()
kernel.add_service(AzureChatCompletion())
# Add any plugins to the kernel that the planner will leverage
kernel.add_plugins(...)

chat_history = ChatHistory()
chat_hitory.add_user_message("Check current UTC time and return current weather in Boston city.")

request_settings = AzureChatPromptExecutionSettings(function_choice_behavior=FunctionChoiceBehavior.Auto())

response = await kernel.invoke_prompt(
    "Check current UTC time and return current weather in Boston city.", 
    arguments=KernelArguments(settings=request_settings),
)
print(response)

Planery nie były dostępne w środowisku SK Java. Użyj funkcji wywołującej bezpośrednio.

Powyższe fragmenty kodu pokazują, jak zmigrować kod korzystający z narzędzia Stepwise Planner do automatycznego wywoływania funkcji. Dowiedz się więcej na temat wywoływania funkcji z ukończeniem czatu.