This repo stores an example of how to train a tensorflow model and deploy to production using Amazon Sagemaker.
- Explore and Train.ipynb - loads the dataset, trains a CNN and saves the model
- sample_utils.py - helper functions used in the notebooks
- tensorflow_serving_container.ipynb - compresses the saved model, uploads it to s3 and deploys the model