Python Client for Cloud AutoML API#

alpha pypi versions

The Cloud AutoML API is a suite of machine learning products that enables developers with limited machine learning expertise to train high-quality models specific to their business needs, by leveraging Google’s state-of-the-art transfer learning, and Neural Architecture Search technology.

Quick Start#

In order to use this library, you first need to go through the following steps:

  1. Select or create a Cloud Platform project.

  2. Enable billing for your project.

  3. Enable the Cloud AutoML API.

  4. Setup Authentication.

Installation#

Install this library in a virtualenv using pip. virtualenv is a tool to create isolated Python environments. The basic problem it addresses is one of dependencies and versions, and indirectly permissions.

With virtualenv, it’s possible to install this library without needing system install permissions, and without clashing with the installed system dependencies.

Supported Python Versions#

Python >= 3.5

Deprecated Python Versions#

Python == 2.7. Python 2.7 support will be removed on January 1, 2020.

Mac/Linux#

pip install virtualenv
virtualenv <your-env>
source <your-env>/bin/activate
<your-env>/bin/pip install google-cloud-automl

Windows#

pip install virtualenv
virtualenv <your-env>
<your-env>\Scripts\activate
<your-env>\Scripts\pip.exe install google-cloud-automl

Example Usage#

from google.cloud.automl_v1beta1 import PredictionServiceClient

client = PredictionServiceClient()
model_path = client.model_path('my-project-123', 'us-central', 'model-name')
payload = {...}
params = {'foo': 1}
response = client.predict(model_path, payload, params=params)

Next Steps#

Making & Testing Local Changes#

If you want to make changes to this library, here is how to set up your development environment:

  1. Make sure you have virtualenv installed and activated as shown above.

  2. Run the following one-time setup (it will be persisted in your virtualenv):

    pip install -r ../docs/requirements.txt
    pip install -U nox mock pytest
    
  3. If you want to run all tests, you will need a billing-enabled GCP project, and a service account with access to the AutoML APIs. Note: the first time the tests run in a new project it will take a _long_ time, on the order of 2-3 hours. This is one-time setup that will be skipped in future runs.

export PROJECT_ID=<project-id> GOOGLE_APPLICATION_CREDENTIALS=</path/to/creds.json>
nox