Quick Start (Model Developers)¶

As a Model Developer, you can manage models and manage train & inference jobs on Rafiki.

This quickstart only highlights the key methods available to manage models. To learn about how to manage train & inference jobs, go to Quick Start (Application Developers). To learn more about what you can do on Rafiki, explore the methods of rafiki.client.Client.

We assume that you have access to a running instance of Rafiki Admin at <rafiki_host>:<admin_port> and Rafiki Admin Web at <rafiki_host>:<admin_web_port>.

Installing the client¶

Install Python 3.6 such that the python and pip point to the correct installation of Python (see Installing Python)
Clone the project at https://github.com/nginyc/rafiki (e.g. with Git)
Within the project’s root folder, install Rafiki Client’s dependencies by running:
pip install -r ./rafiki/requirements.txt

Initializing the client¶

Example:

from rafiki.client import Client
client = Client(admin_host='localhost', admin_port=3000)
client.login(email='model_developer@rafiki', password='rafiki')

See also

rafiki.client.Client.login()

Creating models¶

To create a model, you will need to submit a model class that conforms to the specification by rafiki.model.BaseModel, written in a single Python file. The model’s implementation should conform to a specific task (see Supported Tasks).

Refer to the parameters of rafiki.client.Client.create_model() for configuring how your model runs on Rafiki, and refer to Creating Models to understand more about how to write & test models for Rafiki.

Examples:

client.create_model(
    name='TfFeedForward',
    task='IMAGE_CLASSIFICATION',
    model_file_path='examples/models/image_classification/TfFeedForward.py',
    model_class='TfFeedForward',
    dependencies={ 'tensorflow': '1.12.0' }
)

client.create_model(
    name='SkDt',
    task='IMAGE_CLASSIFICATION',
    model_file_path='examples/models/image_classification/SkDt.py',
    model_class='SkDt',
    dependencies={ 'scikit-learn': '0.20.0' }
)

See also

rafiki.client.Client.create_model()

Listing available models by task¶

Example:

client.get_available_models(task='IMAGE_CLASSIFICATION')

Output:

[{'access_right': 'PRIVATE',
 'datetime_created': 'Mon, 17 Dec 2018 07:06:03 GMT',
 'dependencies': {'tensorflow': '1.12.0'},
 'id': '45df3f34-53d7-4fb8-a7c2-55391ea10030',
 'name': 'TfFeedForward',
 'task': 'IMAGE_CLASSIFICATION',
 'user_id': 'fb5671f1-c673-40e7-b53a-9208eb1ccc50'},
 {'access_right': 'PRIVATE',
 'datetime_created': 'Mon, 17 Dec 2018 07:06:03 GMT',
 'dependencies': {'scikit-learn': '0.20.0'},
 'id': 'd0ea96ce-478b-4167-8a84-eb36ae631235',
 'name': 'SkDt',
 'task': 'IMAGE_CLASSIFICATION',
 'user_id': 'fb5671f1-c673-40e7-b53a-9208eb1ccc50'}]

Deleting a model¶

Example:

client.delete_model('fb5671f1-c673-40e7-b53a-9208eb1ccc50')

See also

rafiki.client.Client.delete_model()