How to instantiate a GreatAI service

The core value of the great-ai library lies in its [great_ai.deploy.GreatAI][] class. In order to take advantage of it, you need to create an instance wrapping your code.

Let's say that you have the following greeter function:

def my_greeter_function(your_name):
    return f'Hi {your_name}!'

You can simply decorate (wrap) this function with the GreatAI.create factory.

from great_ai import GreatAI

@GreatAI.create
def greeter(your_name):
    return f'Hi {your_name}!'

??? info "Why not simply use @GreatAI?" The purpose of the GreatAI.create is simply to provide you with type-checking through MyPy, Pylance, and similar libraries. However, the overloading support for __new__ is lacking in MyPy, thus, a static factory method is used instead.

With types

Type annotating your codebase{ target=_blank } can save you from lots of trivial mistakes, that's why it's highly advised. Simply add the expected types to your function's signature.

from great_ai import GreatAI

@GreatAI.create
def type_safe_greeter(your_name: str) -> str:
    return f'Hi {your_name}!'

This not only allows you to statically typecheck your code, but by default, GreatAI will check it during runtime as well using typeguard{ target=_blank }.

With async

Asynchronous code can result in immense performance gains in certain cases. For example, you might rely on a third-party service, do database access, or call a remote GreatAI instance. In these cases, you can simply make your function async without any other changes.

from great_ai import GreatAI
from asyncio import sleep

@GreatAI.create
async def async_greeter(your_name: str) -> str:
    await sleep(2)  # simulate IO-heavy operation
    return f'Hi {your_name}!'

With decorators

GreatAI can decorate already decorated functions. The only restriction is that @GreatAI.create always have to come last. There are two built-in decorators that you can use to customise your function.

Using `use_model`

If you have previously saved a model with save_model, you can inject it into your function by calling use_model.

from great_ai import GreatAI, use_model

@GreatAI.create
@use_model('name_of_my_model', version='latest') #(1)
def type_safe_greeter(your_name: str, model) -> str:
    return f'Hi {your_name}!'

assert type_safe_greeter('Andras').output == 'Hi Andras'

By default, the parameter named model will be replaced by the loaded model. This behaviour can be customised by setting the model_kwarg_name. This way, even multiple models can be injected into a single function.

!!! important You must call @use_model before GreatAI.create. Feel free to use @use_model in other places of the code base, it works equally well outside of GreatAI services.

Using `parameter`

If you wish to turn of logging or specify custom validation for your parameters, you can use the @parameter decorator.

!!! note By default, all parameters that are not affected by an explicit @parameter or @use_model decorator, are automatically decorated with @parameter when GreatAI.create is called.

from great_ai import GreatAI, use_model

@GreatAI.create
@use_model('name_of_my_model', version='latest')
def type_safe_greeter(your_name: str, model) -> str:
    return f'Hi {your_name}!'

assert type_safe_greeter('Andras').output == 'Hi Andras'

!!! important You must call @parameter before GreatAI.create. Feel free to use @parameter in other places of the code base, it works equally well outside of GreatAI services.

Complex example

Refer to the following example summarising the options you have when instantiating a GreatAI service.

from great_ai import save_model, GreatAI, parameter, use_model

save_model(4, 'secret-number') #(2)

@GreatAI.create
@parameter('positive_number', validator=lambda n: n > 0)
@use_model('secret-number', version='latest', model_kwarg_name='secret')
def add_number(positive_number: int, secret: int) -> int:
    return positive_number + secret

assert add_number(1).output == 5

Refer to storing models for specifying where to store your models.

4.5 KiB Raw Blame History