Introduction

I love flowers. The lotus flower above is one of my most favorite flower photos taken during my visit to the Summer Palace Beijing in 2008. Since I am a developer and enjoy learning and working on artificial intelligence and cloud projects, I decide to write this blog post to share my project on building a real-world flower classifier with TensorFlow, Amazon SageMaker and Docker.

This post shows step-by-step guide on:

• Using ready-to-use Flower dataset from TensorFlow Datasets.
• Using Transfer Learning for feature extraction from a pre-trained model from TensorFlow Hub.
• Using tf.data API to build input pipelines for the dataset split into training, validation and test datasets.
• Using tf.keras API to build, train and evaluate the model.
• Using Callback to define early stopping threshold for model training.
• Preparing training script to train and export the model in SavedModel format for deploy with TensorFlowand Amazon SageMaker Python SDK.
• Preparing inference code and configuration to run the TensorFlow Serving ModelServer for serving the model.
• Building custom Docker Containers for training and serving the TensorFlow model with Amazon SageMaker Python SDK and SageMaker TensorFlow Training Toolkit in Local mode.

#transfer-learning #docker-container #tensorflow-serving #tensorflow2 #amazon-sagemaker