MLAuth

This is a collection of packages that use Google Cloud's ML APIs to act as a form of authentication. Basically, pass in a term or a condition and an ML-able artifact (iamge for vision, audio file for speech, etc..). Then the package uses the appropriate ML API to find the term in the artifact.

Requirements

Create a Google Cloud Project
Create env variables for project:

export mlauth_project=[Name of proejct]
export mlauth_bucket=[Name of testing bucket]

Run init command

make init

This will do the following:

Create a bucket for testing
Copy files for testing to bucket
Create a service account for applications
Create a key file so that we can use this service account in default credentials mode
Activate Services in your project
- Speech
- Vision
- Natural Language

After this is done you need to set the environment variable for api calls to work.

Set the env variable for Google Cloud default credentials

export GOOGLE_APPLICATION_CREDENTIALS=[Project Path]/creds/creds.json

You can run make test afterwards to determine if everything worked.

This is not an official Google product.

tpryan/mlauth

MLAuth

Requirements