From 4e062cb5d31d27894d0e74eea6e49654ff13adc8 Mon Sep 17 00:00:00 2001
From: winstonaws <31297115+winstonaws@users.noreply.github.com>
Date: Tue, 3 Apr 2018 19:41:58 -0700
Subject: [PATCH] Add README description (#20)

---
 README.rst | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/README.rst b/README.rst
index 2cfbc89..178476e 100644
--- a/README.rst
+++ b/README.rst
@@ -2,7 +2,18 @@
 SageMaker Container Support
 ===========================
 
-This support code is used for making machine learning frameworks (e.g. MXNet, TensorFlow) run on Amazon SageMaker.
+This package makes it easier to build docker images which run on Amazon SageMaker. For example, the official SageMaker `TensorFlow <https://github.com/aws/sagemaker-tensorflow-containers>`_ and `MXNet <https://github.com/aws/sagemaker-mxnet-containers>`_ images use this package. The requirements for SageMaker-compatible images are documented here: https://docs.aws.amazon.com/sagemaker/latest/dg/your-algorithms.html
+
+Features provided:
+
+- Integration with `SageMaker Python SDK  <https://github.com/aws/sagemaker-python-sdk>`_ Estimators, including:
+
+  - Downloading user-provided python code
+  - Deserializing hyperparameters (which preserves their python types)
+
+- ``bin/entry.py``, which acts as the docker entrypoint required by SageMaker
+- Reading in the metadata files provided to the container during training, as described here: https://docs.aws.amazon.com/sagemaker/latest/dg/your-algorithms-training-algo.html
+- nginx + Gunicorn HTTP server for serving inference requests which complies with the interface described here: https://docs.aws.amazon.com/sagemaker/latest/dg/your-algorithms-inference-code.html
 
 License
 -------