The entire Cloud AutoML platform will help businesses scale up their AI capabilities without requiring advanced machine learning expertise. AutoML enables the end user to experience in learning high-quality NN models. Cloud AutoML Vision is built on GCP transfer learning and neural architecture search technologies. It is a simple, secure, and flexible ML API. Cloud AutoML provides a simple graphical user interface (GUI) to train, evaluate, improve, and deploy models. It is integrated easily with all other APIs of GCP.
The architecture of AutoML is as follows:
The core components used in the model are:
- Cloud Storage
- Cloud Dataflow
- Cloud Vision API
- Cloud Image API
- REST API
- RPC API
- Cloud Endpoints
- BigQuery/BigTable
The applications of this model include:
- Animation and gaming
- Digital Commerce
- Conservation of animals and their habits (London Zoo)
- Recommendation systems
- Healthcare and medical
- Automobiles
- Home appliances