+ Gradient

How to use Gradient and



Use Jupyter Notebooks to develop and train models on Gradient.

Gradient has a tight integration with Jupyter. Gradient Notebooks are a fully managed cloud environment based on the open-source Jupyter project and compatible with any existing .ipynb notebooks.

A Gradient Notebook gives you access to a full Jupyter Notebook environment. Within the Notebook, you can store an unlimited number of documents and other files. You can think of a Gradient Notebook as a persistent, on-demand workspace in the cloud.

Storing data

Every notebook in your account automatically includes a persistent filesystem. Use this directory to store datasets, model checkpoints, and more.  Learn more here. Gradient also includes a public datasets repository.

Offline mode

One of the advantages of using Gradient Notebooks is that you can view their contents without actually running the Notebook. Just click the open button on any public or private Notebook and it will open a static version. This is handy for exploring public Notebooks or quickly glancing at your work.

Sharing Notebooks

You can generate a link to share your Notebook with friends and colleagues with one click. Public Notebooks can be forked by others into their own account. To learn more about Gradient Notebooks, try forking a public demo notebook here. The ML Showcase includes several working examples of projects you can run with a couple clicks. (And ML Showcase welcomes new project submissions!)

Creating a Jupyter Notebook

When launching a Notebook via the web interface, CLI, or workflow, you can select a pre-built template or pass in a custom Docker image path (e.g. <inline-code>ermaker/keras-jupyter<inline-code>. Pre-built templates are updated regularly and optimized to run on Gradient.  Any Docker container is supported on the Gradient platform so customizing your own image is simple.

A set of pre-built containers can be used as a starting point within Gradient

When using the CLI, the command would like something like this: