main

This example shows how to do image classification from scratch using model with BatchNorm layer, without leveraging pre-trained weights or a pre-made model. We demonstrate the workflow on the FashionMnist classification dataset.

It includes:

• dataset loading from S3

• model compilation

• model training

• model evaluation

This examples demonstrates the inference concept:

• Weights are loaded from .h5 file, configuration is loaded from .json file.

• Model is evaluated after loading to obtain accuracy value.

• No additional training.

• No new layers are added.

NOTE: Model and weights are resources in api module.

We demonstrate the workflow on the FashionMnist classification dataset.