Training Mobile-Ready models using TensorFlow Lite¶

Table of Contents¶

• Introduction
• Prerequisites
• About TensorFlow Lite
• Supported Models
• Set Environment Variables
• Imports
• Data preparation
• Visualize Training data
• Train Model
  • Model instantiation options
  • Visualizing a few results
  • Save Model to an intermediate format
  • Load a model and Train it further
• Exporting model to mobile ready format
• Deployment

Introduction¶

We can use arcgis.learn to train deep learning models that can be deployed on mobile devices using ArcGIS field apps. This enables AI on the edge and the simplification of field workers' jobs. This is done using TensorFlow which enables us to save trained models to '.tflite' format. A few applications involved in this workflow are

• Assisted intelligent survey
• Disconnected Decision support
• Automatic feature extraction from street view
• Automated Rapid data collection on the move