Functions Facade

It is possible to manage the models of a ML Model Manager server trough functions rather than trough an object Client.

The library provides a set of functions that expose the same interface as the client object and allow operations to be performed in a single call.

Let’s overview the functions.

add_model(...)

Allows you to create a model.

>>> from devo_ml.modelmanager import add_model
>>>
>>> add_model(
...     "http://localhost",                         # url
...     "8a3vf98ai28sar1234lkj2l43td6f89a",         # token
...     "pokemon_onnx_regression",                  # model name
...     "ONNX",                                     # model engine
...     "~/models/pokemon.onnx",                    # model file
...     description="A funny Pokemon prediction"
... )
>>>

If the model you are trying to add already exists in the system, a ModelAlreadyExists error will be thrown.

Trough the force parameter you can manage whether to override the model.

get_model(...)

Allows you to retrieve model.

>>> from devo_ml.modelmanager import get_model
>>>
>>> get_model(
...     "http://localhost",                         # url
...     "8a3vf98ai28sar1234lkj2l43td6f89a",         # token
...     "pokemon_onnx_regression",                  # model name
...     download_path="~/models",
... )
{
   'id': 35,
   'name': 'pokemon_onnx_regression',
   'engine': 'ONNX',
   'description': 'A funny Pokemon prediction',
   ...
}
>>>

The parameter download_path indicates that we want download the model file, and where we want to download the model file. Omitting this parameter will not download file. Note that you can not use a custom downloader with the function.

If the model you are trying to query does not exists, a ModelNotFound error will be thrown.

find_model(...)

It behaves the same as get_model(…) except it returns None instead of throw an error if the model doesn’t exists. It is a convenient way to get the model data without the need of catching errors.

>>> from devo_ml.modelmanager import find_model
>>>
>>> find_model(
...     "http://localhost",                         # url
...     "8a3vf98ai28sar1234lkj2l43td6f89a",         # token
...     "pokemon_onnx_regression",                  # model name
...     download_path="~/models",
... )
{
   'id': 35,
   'name': 'pokemon_onnx_regression',
   'engine': 'ONNX',
   'description': 'A funny Pokemon prediction',
   ...
}
>>>

get_models(...)

Allows you retrieve a list of all the models in the system. Note that it doesn’t allow downloading model files.

>>> from devo_ml.modelmanager import get_models
>>>
>>> get_models(
...     "http://localhost",                         # url
...     "8a3vf98ai28sar1234lkj2l43td6f89a",         # token
... )
[
    {
        'id': 35,
        'name': 'pokemon_onnx_regression',
        'engine': 'ONNX',
        'description': 'A funny Pokemon prediction',
        ...
    }, {
        'id': 36,
        'name': 'credit_card_gjp',
        'engine': 'H2O',
        'description': 'gjp model on credit card fraud dataset',
        ...
    }, {
    ...
]
>>>

Note

You can not use your own downloaders with the functions facade.

Note

You can choose the authentication to use with the auth_type parameter and tune the underlying request with keywords. This is valid for all functions facade.