Using TensorFlow Enterprise with AI Platform Notebooks

This page describes how to get started using TensorFlow Enterprise with AI Platform Notebooks.

In this example, you create a TensorFlow Enterprise instance of AI Platform Notebooks, open a JupyterLab notebook, and run a classification tutorial on using neural networks with Keras.

Before you begin

Before you can use AI Platform Notebooks, you must have a Google Cloud project and enable the AI Platform Notebooks API for that project.
  1. 登录您的 Google 帐号。

    如果您还没有 Google 帐号,请注册一个新帐号

  2. 在 Cloud Console 的项目选择器页面上,选择或创建 Cloud 项目。


  3. 确保您的 Google Cloud 项目已启用结算功能。 了解如何确认您的项目已启用结算功能

  4. 启用 AI Platform Notebooks API。

    启用 API

Create an AI Platform Notebooks instance

To create a TensorFlow Enterprise AI Platform Notebooks instance, complete these steps:

  1. Go to the AI Platform Notebooks page in the Google Cloud Console.

    Go to the AI Platform Notebooks page

  2. Click New Instance, select TensorFlow Enterprise 1.15, and then select Without GPUs .

  3. Click Create.

  4. AI Platform Notebooks creates a new instance based on your selected framework. An Open JupyterLab link becomes active when it's ready to use.

Open the notebook

Complete these steps to open a notebook instance:
  1. On the AI Platform Notebooks page in the Google Cloud Console, click Open JupyterLab to open the notebook.

  2. AI Platform Notebooks opens your notebook.

Run a classification tutorial in your notebook instance

Complete these steps to try out your new notebook by running a classification tutorial:

  1. In your JupyterLab notebook, on the left, double-click the tutorials folder to open it, and navigate to and open tutorials/keras/basic_classification.ipynb.

  2. Click the run button to run cells of the tutorial.

What's next