Exciting news! Gradient has launched a FREE GPU plan. Read More
Project Details

Gradient Python SDK end-to-end example

Train and deploy a model with the Gradient SDK using the classic MNIST handwritten digits dataset and TensorFlow

The Gradient Team


This Notebook provides a step-by-step walkthrough of the Gradient Python SDK.  This library allows you to programmatically interact with Gradient from within a Jupyter Notebook environment (like this tutorial) or from within any python project. The Gradient Python SDK supplements the Gradient CLI functionality and UI with the added ability to automate actions and pipelines.

In this example, we'll be training a convolutional neural network to recognize handwritten digits using the classic MNIST dataset and TensorFlow.  This demo walks through training a model, storing it in Gradient, deploying it as a RESTful API endpoint, and then making a prediction.  The purpose of the tutorial is to demonstrate how simple it is to build a machine learning pipeline from start to finish.  The code in this example lives in this GitHub repo: https://github.com/Paperspace/mnist-sample.git

View the blog announcement post here https://blog.paperspace.com/new-gradient-sdk/