# Use Amazon SageMaker Studio Classic Notebooks
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
Amazon SageMaker Studio Classic notebooks are collaborative notebooks that you can launch quickly because you don't need to set up compute instances and file storage beforehand. Studio Classic notebooks provide persistent storage, which enables you to view and share notebooks even if the instances that the notebooks run on are shut down.
You can share your notebooks with others, so that they can easily reproduce your results and collaborate while building models and exploring your data. You provide access to a read-only copy of the notebook through a secure URL. Dependencies for your notebook are included in the notebook's metadata. When your colleagues copy the notebook, it opens in the same environment as the original notebook.
A Studio Classic notebook runs in an environment defined by the following:
+ Amazon EC2 instance type – The hardware configuration the notebook runs on. The configuration includes the number and type of processors (vCPU and GPU), and the amount and type of memory. The instance type determines the pricing rate.
+ SageMaker image – A container image that is compatible with SageMaker Studio Classic. The image consists of the kernels, language packages, and other files required to run a notebook in Studio Classic. There can be multiple images in an instance. For more information, see [Custom Images in Amazon SageMaker Studio Classic](studio-byoi.md).
+ KernelGateway app – A SageMaker image runs as a KernelGateway app. The app provides access to the kernels in the image. There is a one-to-one correspondence between a SageMaker AI image and a KernelGateway app.
+ Kernel – The process that inspects and runs the code contained in the notebook. A kernel is defined by a *kernel spec* in the image. There can be multiple kernels in an image.
You can change any of these resources from within the notebook.
The following diagram outlines how a notebook kernel runs in relation to the KernelGateway App, User, and domain.
![\[How a notebook kernel runs in relation to the KernelGateway App, User, and domain.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/studio-components.png)
Sample SageMaker Studio Classic notebooks are available in the [aws\$1sagemaker\$1studio](https://github.com/awslabs/amazon-sagemaker-examples/tree/master/aws_sagemaker_studio) folder of the [Amazon SageMaker example GitHub repository](https://github.com/awslabs/amazon-sagemaker-examples). Each notebook comes with the necessary SageMaker image that opens the notebook with the appropriate kernel.
We recommend that you familiarize yourself with the SageMaker Studio Classic interface and the Studio Classic notebook toolbar before creating or using a Studio Classic notebook. For more information, see [Amazon SageMaker Studio Classic UI Overview](studio-ui.md) and [Use the Studio Classic Notebook Toolbar](notebooks-menu.md).
**Topics**
+ [How Are Amazon SageMaker Studio Classic Notebooks Different from Notebook Instances?](notebooks-comparison.md)
+ [Get Started with Amazon SageMaker Studio Classic Notebooks](notebooks-get-started.md)
+ [Amazon SageMaker Studio Classic Tour](gs-studio-end-to-end.md)
+ [Create or Open an Amazon SageMaker Studio Classic Notebook](notebooks-create-open.md)
+ [Use the Studio Classic Notebook Toolbar](notebooks-menu.md)
+ [Install External Libraries and Kernels in Amazon SageMaker Studio Classic](studio-notebooks-add-external.md)
+ [Share and Use an Amazon SageMaker Studio Classic Notebook](notebooks-sharing.md)
+ [Get Amazon SageMaker Studio Classic Notebook and App Metadata](notebooks-run-and-manage-metadata.md)
+ [Get Notebook Differences in Amazon SageMaker Studio Classic](notebooks-diff.md)
+ [Manage Resources for Amazon SageMaker Studio Classic Notebooks](notebooks-run-and-manage.md)
+ [Usage Metering for Amazon SageMaker Studio Classic Notebooks](notebooks-usage-metering.md)
+ [Available Resources for Amazon SageMaker Studio Classic Notebooks](notebooks-resources.md)
# How Are Amazon SageMaker Studio Classic Notebooks Different from Notebook Instances?
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
When you're starting a new notebook, we recommend that you create the notebook in Amazon SageMaker Studio Classic instead of launching a notebook instance from the Amazon SageMaker AI console. There are many benefits to using a Studio Classic notebook, including the following:
+ **Faster: **Starting a Studio Classic notebook is faster than launching an instance-based notebook. Typically, it is 5-10 times faster than instance-based notebooks.
+ **Easy notebook sharing: **Notebook sharing is an integrated feature in Studio Classic. Users can generate a shareable link that reproduces the notebook code and also the SageMaker image required to execute it, in just a few clicks.
+ **Latest Python SDK: **Studio Classic notebooks come pre-installed with the latest [Amazon SageMaker Python SDK](https://sagemaker.readthedocs.io/en/stable).
+ **Access all Studio Classic features: **Studio Classic notebooks are accessed from within Studio Classic. This enables you to build, train, debug, track, and monitor your models without leaving Studio Classic.
+ **Persistent user directories:** Each member of a Studio team gets their own home directory to store their notebooks and other files. The directory is automatically mounted onto all instances and kernels as they're started, so their notebooks and other files are always available. The home directories are stored in Amazon Elastic File System (Amazon EFS) so that you can access them from other services.
+ **Direct access:** When using IAM Identity Center, you use your IAM Identity Center credentials through a unique URL to directly access Studio Classic. You don't have to interact with the AWS Management Console to run your notebooks.
+ **Optimized images:** Studio Classic notebooks are equipped with a set of predefined SageMaker image settings to get you started faster.
**Note**
Studio Classic notebooks don't support *local mode*. However, you can use a notebook instance to train a sample of your dataset locally, and then use the same code in a Studio Classic notebook to train on the full dataset.
When you open a notebook in SageMaker Studio Classic, the view is an extension of the JupyterLab interface. The primary features are the same, so you'll find the typical features of a Jupyter notebook and JupyterLab. For more information about the Studio Classic interface, see [Amazon SageMaker Studio Classic UI Overview](studio-ui.md).
# Get Started with Amazon SageMaker Studio Classic Notebooks
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
To get started, you or your organization's administrator need to complete the SageMaker AI domain onboarding process. For more information, see [Amazon SageMaker AI domain overview](gs-studio-onboard.md).
You can access a Studio Classic notebook in any of the following ways:
+ You receive an email invitation to access Studio Classic through your organization's IAM Identity Center, which includes a direct link to login to Studio Classic without having to use the Amazon SageMaker AI console. You can proceed to the [Next Steps](#notebooks-get-started-next-steps).
+ You receive a link to a shared Studio Classic notebook, which includes a direct link to log in to Studio Classic without having to use the SageMaker AI console. You can proceed to the [Next Steps](#notebooks-get-started-next-steps).
+ You onboard to a domain and then log in to the SageMaker AI console. For more information, see [Amazon SageMaker AI domain overview](gs-studio-onboard.md).
## Launch Amazon SageMaker AI
Complete the steps in [Launch Amazon SageMaker Studio Classic](studio-launch.md) to launch Studio Classic.
## Next Steps
Now that you're in Studio Classic, you can try any of the following options:
+ To create a Studio Classic notebook or explore Studio Classic end-to-end tutorial notebooks – See [Amazon SageMaker Studio Classic Tour](gs-studio-end-to-end.md) in the next section.
+ To familiarize yourself with the Studio Classic interface – See [Amazon SageMaker Studio Classic UI Overview](studio-ui.md) or try the **Getting started notebook** by selecting **Open the Getting started notebook** in the **Quick actions** section of the Studio Classic Home page.
# Amazon SageMaker Studio Classic Tour
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
For a walkthrough that takes you on a tour of the main features of Amazon SageMaker Studio Classic, see the [xgboost\$1customer\$1churn\$1studio.ipynb](https://sagemaker-examples.readthedocs.io/en/latest/aws_sagemaker_studio/getting_started/xgboost_customer_churn_studio.html) sample notebook from the [aws/amazon-sagemaker-examples](https://github.com/aws/amazon-sagemaker-examples) GitHub repository. The code in the notebook trains multiple models and sets up the SageMaker Debugger and SageMaker Model Monitor. The walkthrough shows you how to view the trials, compare the resulting models, show the debugger results, and deploy the best model using the Studio Classic UI. You don't need to understand the code to follow this walkthrough.
**Prerequisites**
To run the notebook for this tour, you need:
+ An IAM account to sign in to Studio. For information, see [Amazon SageMaker AI domain overview](gs-studio-onboard.md).
+ Basic familiarity with the Studio user interface and Jupyter notebooks. For information, see [Amazon SageMaker Studio Classic UI Overview](studio-ui.md).
+ A copy of the [aws/amazon-sagemaker-examples](https://github.com/aws/amazon-sagemaker-examples) repository in your Studio environment.
**To clone the repository**
1. Launch Studio Classic following the steps in [Launch Amazon SageMaker Studio Classic](studio-launch.md) For users in IAM Identity Center, sign in using the URL from your invitation email.
1. On the top menu, choose **File**, then **New**, then **Terminal**.
1. At the command prompt, run the following command to clone the [aws/amazon-sagemaker-examples](https://github.com/aws/amazon-sagemaker-examples) GitHub repository.
```
$ git clone https://github.com/aws/amazon-sagemaker-examples.git
```
**To navigate to the sample notebook**
1. From the **File Browser** on the left menu, select **amazon-sagemaker-examples**.
1. Navigate to the example notebook with the following path.
`~/amazon-sagemaker-examples/aws_sagemaker_studio/getting_started/xgboost_customer_churn_studio.ipynb`
1. Follow the notebook to learn about Studio Classic's main features.
**Note**
If you encounter an error when you run the sample notebook, and some time has passed from when you cloned the repository, review the notebook on the remote repository for updates.
# Create or Open an Amazon SageMaker Studio Classic Notebook
**Important**
Custom IAM policies that allow Amazon SageMaker Studio or Amazon SageMaker Studio Classic to create Amazon SageMaker resources must also grant permissions to add tags to those resources. The permission to add tags to resources is required because Studio and Studio Classic automatically tag any resources they create. If an IAM policy allows Studio and Studio Classic to create resources but does not allow tagging, "AccessDenied" errors can occur when trying to create resources. For more information, see [Provide permissions for tagging SageMaker AI resources](security_iam_id-based-policy-examples.md#grant-tagging-permissions).
[AWS managed policies for Amazon SageMaker AI](security-iam-awsmanpol.md) that give permissions to create SageMaker resources already include permissions to add tags while creating those resources.
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
When you [Create a Notebook from the File Menu](#notebooks-create-file-menu) in Amazon SageMaker Studio Classic or [Open a notebook in Studio Classic](#notebooks-open) for the first time, you are prompted to set up your environment by choosing a SageMaker image, a kernel, an instance type, and, optionally, a lifecycle configuration script that runs on image start-up. SageMaker AI launches the notebook on an instance of the chosen type. By default, the instance type is set to `ml.t3.medium` (available as part of the [AWS Free Tier](https://aws.amazon.com/free)) for CPU-based images. For GPU-based images, the default instance type is `ml.g4dn.xlarge`.
If you create or open additional notebooks that use the same instance type, whether or not the notebooks use the same kernel, the notebooks run on the same instance of that instance type.
After you launch a notebook, you can change its instance type, SageMaker image, and kernel from within the notebook. For more information, see [Change the Instance Type for an Amazon SageMaker Studio Classic Notebook](notebooks-run-and-manage-switch-instance-type.md) and [Change the Image or a Kernel for an Amazon SageMaker Studio Classic Notebook](notebooks-run-and-manage-change-image.md).
**Note**
You can have only one instance of each instance type. Each instance can have multiple SageMaker images running on it. Each SageMaker image can run multiple kernels or terminal instances.
Billing occurs per instance and starts when the first instance of a given instance type is launched. If you want to create or open a notebook without the risk of incurring charges, open the notebook from the **File** menu and choose **No Kernel** from the **Select Kernel** dialog box. You can read and edit a notebook without a running kernel but you can't run cells.
Billing ends when the SageMaker image for the instance is shut down. For more information, see [Usage Metering for Amazon SageMaker Studio Classic Notebooks](notebooks-usage-metering.md).
For information about shutting down the notebook, see [Shut down resources](notebooks-run-and-manage-shut-down.md#notebooks-run-and-manage-shut-down-sessions).
**Topics**
+ [Open a notebook in Studio Classic](#notebooks-open)
+ [Create a Notebook from the File Menu](#notebooks-create-file-menu)
+ [Create a Notebook from the Launcher](#notebooks-create-launcher)
+ [List of the available instance types, images, and kernels](#notebooks-instance-image-kernels)
## Open a notebook in Studio Classic
Amazon SageMaker Studio Classic can only open notebooks listed in the Studio Classic file browser. For instructions on uploading a notebook to the file browser, see [Upload Files to Amazon SageMaker Studio Classic](studio-tasks-files.md) or [Clone a Git Repository in Amazon SageMaker Studio Classic](studio-tasks-git.md).
**To open a notebook**
1. In the left sidebar, choose the **File Browser** icon ( ![\[Black square icon representing a placeholder or empty image.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/folder.png)) to display the file browser.
1. Browse to a notebook file and double-click it to open the notebook in a new tab.
## Create a Notebook from the File Menu
**To create a notebook from the File menu**
1. From the Studio Classic menu, choose **File**, choose **New**, and then choose **Notebook**.
1. In the **Change environment** dialog box, use the dropdown menus to select your **Image**, **Kernel**, **Instance type**, and **Start-up script**, then choose **Select**. Your notebook launches and opens in a new Studio Classic tab.
![\[Studio Classic notebook environment setup.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/studio-notebook-environment-setup.png)
## Create a Notebook from the Launcher
**To create a notebook from the Launcher**
1. To open the Launcher, choose **Amazon SageMaker Studio Classic** at the top left of the Studio Classic interface or use the keyboard shortcut `Ctrl + Shift + L`.
To learn about all the available ways to open the Launcher, see [Use the Amazon SageMaker Studio Classic Launcher](studio-launcher.md)
1. In the Launcher, in the **Notebooks and compute resources** section, choose **Change environment**.
![\[SageMaker Studio Classic set notebook environment.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/studio-launcher-notebook-creation.png)
1. In the **Change environment** dialog box, use the dropdown menus to select your **Image**, **Kernel**, **Instance type**, and **Start-up script**, then choose **Select**.
1. In the Launcher, choose **Create notebook**. Your notebook launches and opens in a new Studio Classic tab.
To view the notebook's kernel session, in the left sidebar, choose the **Running Terminals and Kernels** icon (![\[Black square icon representing a placeholder or empty image.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/running-terminals-kernels.png)). You can stop the notebook's kernel session from this view.
## List of the available instance types, images, and kernels
For a list of all available resources, see:
+ [Instance Types Available for Use With Amazon SageMaker Studio Classic Notebooks](notebooks-available-instance-types.md)
+ [Amazon SageMaker Images Available for Use With Studio Classic Notebooks](notebooks-available-images.md)
# Use the Studio Classic Notebook Toolbar
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
Amazon SageMaker Studio Classic notebooks extend the JupyterLab interface. For an overview of the original JupyterLab interface, see [The JupyterLab Interface](https://jupyterlab.readthedocs.io/en/latest/user/interface.html).
The following image shows the toolbar and an empty cell from a Studio Classic notebook.
![\[SageMaker Studio Classic notebook menu.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/studio-notebook-menu.png)
When you pause on a toolbar icon, a tooltip displays the icon function. Additional notebook commands are found in the Studio Classic main menu. The toolbar includes the following icons:
| Icon | Description |
| --- | --- |
| ![\[The Save and checkpoint icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-save-and-checkpoint.png) | **Save and checkpoint** Saves the notebook and updates the checkpoint file. For more information, see [Get the Difference Between the Last Checkpoint](notebooks-diff.md#notebooks-diff-checkpoint). |
| ![\[The Insert cell icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-insert-cell.png) | **Insert cell** Inserts a code cell below the current cell. The current cell is noted by the blue vertical marker in the left margin. |
| ![\[The Cut, copy, and paste cells icons.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-cut-copy-paste.png) | **Cut, copy, and paste cells** Cuts, copies, and pastes the selected cells. |
| ![\[The Run cells icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-run.png) | **Run cells** Runs the selected cells and then makes the cell that follows the last selected cell the new selected cell. |
| ![\[The Interrupt kernel icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-interrupt-kernel.png) | **Interrupt kernel** Interrupts the kernel, which cancels the currently running operation. The kernel remains active. |
| ![\[The Restart kernel icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-restart-kernel.png) | **Restart kernel** Restarts the kernel. Variables are reset. Unsaved information is not affected. |
| ![\[The Restart kernel and run all cells icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-restart-kernel-run-all-cells.png) | **Restart kernel and run all cells** Restarts the kernel, then run all the cells of the notebook. |
| ![\[The Cell type icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-cell-type.png) | **Cell type** Displays or changes the current cell type. The cell types are: [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/sagemaker/latest/dg/notebooks-menu.html) |
| ![\[The Launch terminal icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-launch-terminal.png) | **Launch terminal** Launches a terminal in the SageMaker image hosting the notebook. For an example, see [Get App Metadata](notebooks-run-and-manage-metadata.md#notebooks-run-and-manage-metadata-app). |
| ![\[The Checkpoint diff icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-checkpoint-diff.png) | **Checkpoint diff** Opens a new tab that displays the difference between the notebook and the checkpoint file. For more information, see [Get the Difference Between the Last Checkpoint](notebooks-diff.md#notebooks-diff-checkpoint). |
| ![\[The Git diff icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-git-diff.png) | **Git diff** Only enabled if the notebook is opened from a Git repository. Opens a new tab that displays the difference between the notebook and the last Git commit. For more information, see [Get the Difference Between the Last Commit](notebooks-diff.md#notebooks-diff-git). |
| **2 vCPU \$1 4 GiB** | **Instance type** Displays or changes the instance type the notebook runs in. The format is as follows: `number of vCPUs + amount of memory + number of GPUs` `Unknown` indicates the notebook was opened without specifying a kernel. The notebook runs on the SageMaker Studio instance and doesn't accrue runtime charges. You can't assign the notebook to an instance type. You must specify a kernel and then Studio assigns the notebook to a default type. For more information, see [Create or Open an Amazon SageMaker Studio Classic Notebook](notebooks-create-open.md) and [Change the Instance Type for an Amazon SageMaker Studio Classic Notebook](notebooks-run-and-manage-switch-instance-type.md). |
| ![\[The Cluster icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-cluster.png) | **Cluster** Connect your notebook to an Amazon EMR cluster and scale your ETL jobs or run large-scale model training using Apache Spark, Hive, or Presto. For more information, see [Data preparation using Amazon EMR](studio-notebooks-emr-cluster.md). |
| **Python 3 (Data Science)** | **Kernel and SageMaker Image** Displays or changes the kernel that processes the cells in the notebook. The format is as follows: `Kernel (SageMaker Image)` `No Kernel` indicates the notebook was opened without specifying a kernel. You can edit the notebook but you can't run any cells. For more information, see [Change the Image or a Kernel for an Amazon SageMaker Studio Classic Notebook](notebooks-run-and-manage-change-image.md). |
| ![\[The Kernel busy status icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-kernel-status.png) | **Kernel busy status** Displays the busy status of the kernel. When the edge of the circle and its interior are the same color, the kernel is busy. The kernel is busy when it is starting and when it is processing cells. Additional kernel states are displayed in the status bar at the bottom-left corner of SageMaker Studio. |
| ![\[The Share notebook icon.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-share.png) | **Share notebook** Shares the notebook. For more information, see [Share and Use an Amazon SageMaker Studio Classic Notebook](notebooks-sharing.md). |
To select multiple cells, click in the left margin outside of a cell. Hold down the `Shift` key and use `K` or the `Up` key to select previous cells, or use `J` or the `Down` key to select following cells.
# Install External Libraries and Kernels in Amazon SageMaker Studio Classic
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
Amazon SageMaker Studio Classic notebooks come with multiple images already installed. These images contain kernels and Python packages including scikit-learn, Pandas, NumPy, TensorFlow, PyTorch, and MXNet. You can also install your own images that contain your choice of packages and kernels. For more information on installing your own image, see [Custom Images in Amazon SageMaker Studio Classic](studio-byoi.md).
The different Jupyter kernels in Amazon SageMaker Studio Classic notebooks are separate conda environments. For information about conda environments, see [Managing environments](https://conda.io/docs/user-guide/tasks/manage-environments.html).
## Package installation tools
**Important**
Currently, all packages in Amazon SageMaker notebooks are licensed for use with Amazon SageMaker AI and do not require additional commercial licenses. However, this might be subject to change in the future, and we recommend reviewing the licensing terms regularly for any updates.
The method that you use to install Python packages from the terminal differs depending on the image. Studio Classic supports the following package installation tools:
+ **Notebooks** – The following commands are supported. If one of the following does not work on your image, try the other one.
+ `%conda install`
+ `%pip install`
+ **The Jupyter terminal** – You can install packages using pip and conda directly. You can also use `apt-get install` to install system packages from the terminal.
**Note**
We do not recommend using `pip install -u` or `pip install --user`, because those commands install packages on the user's Amazon EFS volume and can potentially block JupyterServer app restarts. Instead, use a lifecycle configuration to reinstall the required packages on app restarts as shown in [Install packages using lifecycle configurations](#nbi-add-external-lcc).
We recommend using `%pip` and `%conda` to install packages from within a notebook because they correctly take into account the active environment or interpreter being used. For more information, see [Add %pip and %conda magic functions](https://github.com/ipython/ipython/pull/11524). You can also use the system command syntax (lines starting with \$1) to install packages. For example, `!pip install` and `!conda install`.
### Conda
Conda is an open source package management system and environment management system that can install packages and their dependencies. SageMaker AI supports using conda with the conda-forge channel. For more information, see [Conda channels](https://docs.conda.io/projects/conda/en/latest/user-guide/concepts/channels.html). The conda-forge channel is a community channel where contributors can upload packages.
**Note**
Installing packages from conda-forge can take up to 10 minutes. Timing relates to how conda resolves the dependency graph.
All of the SageMaker AI provided environments are functional. User installed packages may not function correctly.
Conda has two methods for activating environments: `conda activate`, and `source activate`. For more information, see [Managing environment](https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html).
**Supported conda operations**
+ `conda install` of a package in a single environment
+ `conda install` of a package in all environments
+ Installing a package from the main conda repository
+ Installing a package from conda-forge
+ Changing the conda install location to use Amazon EBS
+ Supporting both `conda activate` and `source activate`
### Pip
Pip is the tool for installing and managing Python packages. Pip searches for packages on the Python Package Index (PyPI) by default. Unlike conda, pip doesn't have built in environment support. Therfore, pip isn't as thorough as conda when it comes to packages with native or system library dependencies. Pip can be used to install packages in conda environments. You can use alternative package repositories with pip instead of the PyPI.
**Supported pip operations**
+ Using pip to install a package without an active conda environment
+ Using pip to install a package in a conda environment
+ Using pip to install a package in all conda environments
+ Changing the pip install location to use Amazon EBS
+ Using an alternative repository to install packages with pip
### Unsupported
SageMaker AI aims to support as many package installation operations as possible. However, if the packages were installed by SageMaker AI and you use the following operations on these packages, it might make your environment unstable:
+ Uninstalling
+ Downgrading
+ Upgrading
Due to potential issues with network conditions or configurations, or the availability of conda or PyPi, packages may not install in a fixed or deterministic amount of time.
**Note**
Attempting to install a package in an environment with incompatible dependencies can result in a failure. If issues occur, you can contact the library maintainer about updating the package dependencies. When you modify the environment, such as removing or updating existing packages, this may result in instability of that environment.
## Install packages using lifecycle configurations
Install custom images and kernels on the Studio Classic instance's Amazon EBS volume so that they persist when you stop and restart the notebook, and that any external libraries you install are not updated by SageMaker AI. To do that, use a lifecycle configuration that includes both a script that runs when you create the notebook (`on-create)` and a script that runs each time you restart the notebook (`on-start`). For more information about using lifecycle configurations with Studio Classic, see [Use Lifecycle Configurations to Customize Amazon SageMaker Studio Classic](studio-lcc.md). For sample lifecycle configuration scripts, see [SageMaker AI Studio Classic Lifecycle Configuration Samples](https://github.com/aws-samples/sagemaker-studio-lifecycle-config-examples).
# Share and Use an Amazon SageMaker Studio Classic Notebook
**Important**
Custom IAM policies that allow Amazon SageMaker Studio or Amazon SageMaker Studio Classic to create Amazon SageMaker resources must also grant permissions to add tags to those resources. The permission to add tags to resources is required because Studio and Studio Classic automatically tag any resources they create. If an IAM policy allows Studio and Studio Classic to create resources but does not allow tagging, "AccessDenied" errors can occur when trying to create resources. For more information, see [Provide permissions for tagging SageMaker AI resources](security_iam_id-based-policy-examples.md#grant-tagging-permissions).
[AWS managed policies for Amazon SageMaker AI](security-iam-awsmanpol.md) that give permissions to create SageMaker resources already include permissions to add tags while creating those resources.
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
You can share your Amazon SageMaker Studio Classic notebooks with your colleagues. The shared notebook is a copy. After you share your notebook, any changes you make to your original notebook aren't reflected in the shared notebook and any changes your colleague's make in their shared copies of the notebook aren't reflected in your original notebook. If you want to share your latest version, you must create a new snapshot and then share it.
**Topics**
+ [Share a Notebook](#notebooks-sharing-share)
+ [Use a Shared Notebook](#notebooks-sharing-using)
+ [Shared spaces and realtime collaboration](#notebooks-sharing-rtc)
## Share a Notebook
The following screenshot shows the menu from a Studio Classic notebook.
![\[The location of the Share icon in a Studio Classic notebook.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/studio-notebook-menu-share.png)
**To share a notebook**
1. In the upper-right corner of the notebook, choose **Share**.
1. (Optional) In **Create shareable snapshot**, choose any of the following items:
+ **Include Git repo information** – Includes a link to the Git repository that contains the notebook. This enables you and your colleague to collaborate and contribute to the same Git repository.
+ **Include output** – Includes all notebook output that has been saved.
**Note**
If you're an user in IAM Identity Center and you don't see these options, your IAM Identity Center administrator probably disabled the feature. Contact your administrator.
1. Choose **Create**.
1. After the snapshot is created, choose **Copy link** and then choose **Close**.
1. Share the link with your colleague.
After selecting your sharing options, you are provided with a URL. You can share this link with users that have access to Amazon SageMaker Studio Classic. When the user opens the URL, they're prompted to log in using IAM Identity Center or IAM authentication. This shared notebook becomes a copy, so changes made by the recipient will not be reproduced in your original notebook.
## Use a Shared Notebook
You use a shared notebook in the same way you would with a notebook that you created yourself. You must first login to your account, then open the shared link. If you don't have an active session, you receive an error.
When you choose a link to a shared notebook for the first time, a read-only version of the notebook opens. To edit the shared notebook, choose **Create a Copy**. This copies the shared notebook to your personal storage.
The copied notebook launches on an instance of the instance type and SageMaker image that the notebook was using when the sender shared it. If you aren't currently running an instance of the instance type, a new instance is started. Customization to the SageMaker image isn't shared. You can also inspect the notebook snapshot by choosing **Snapshot Details**.
The following are some important considerations about sharing and authentication:
+ If you have an active session, you see a read-only view of the notebook until you choose **Create a Copy**.
+ If you don't have an active session, you need to log in.
+ If you use IAM to login, after you login, select your user profile then choose **Open Studio Classic**. Then you need to choose the link you were sent.
+ If you use IAM Identity Center to login, after you login the shared notebook is opened automatically in Studio.
## Shared spaces and realtime collaboration
A shared space consists of a shared JupyterServer application and a shared directory. A key benefit of a shared space is that it facilitates collaboration between members of the shared space in real time. Users collaborating in a workspace get access to a shared Studio Classic application where they can access, read, and edit their notebooks in real time. Real time collaboration is only supported for JupyterServer applications within a shared space. Users with access to a shared space can simultaneously open, view, edit, and execute Jupyter notebooks in the shared Studio Classic application in that space. For more information about shared spaced and real time collaboration, see [Collaboration with shared spaces](domain-space.md).
# Get Amazon SageMaker Studio Classic Notebook and App Metadata
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
You can access notebook metadata and App metadata using the Amazon SageMaker Studio Classic UI.
**Topics**
+ [Get Studio Classic Notebook Metadata](#notebooks-run-and-manage-metadata-notebook)
+ [Get App Metadata](#notebooks-run-and-manage-metadata-app)
## Get Studio Classic Notebook Metadata
Jupyter notebooks contain optional metadata that you can access through the Amazon SageMaker Studio Classic UI.
**To view the notebook metadata:**
1. In the right sidebar, choose the **Property Inspector** icon (![\[Black square icon representing a placeholder or empty image.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/gears.png)).
1. Open the **Advanced Tools** section.
The metadata should look similar to the following.
```
{
"instance_type": "ml.t3.medium",
"kernelspec": {
"display_name": "Python 3 (Data Science)",
"language": "python",
"name": "python3__SAGEMAKER_INTERNAL__arn:aws:sagemaker:us-west-2::image/datascience-1.0"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.7.10"
}
}
```
## Get App Metadata
When you create a notebook in Amazon SageMaker Studio Classic, the App metadata is written to a file named `resource-metadata.json` in the folder `/opt/ml/metadata/`. You can get the App metadata by opening an Image terminal from within the notebook. The metadata gives you the following information, which includes the SageMaker image and instance type the notebook runs in:
+ **AppType** – `KernelGateway`
+ **DomainId** – Same as the Studio ClassicID
+ **UserProfileName** – The profile name of the current user
+ **ResourceArn** – The Amazon Resource Name (ARN) of the App, which includes the instance type
+ **ResourceName** – The name of the SageMaker image
Additional metadata might be included for internal use by Studio Classic and is subject to change.
**To get the App metadata**
1. In the center of the notebook menu, choose the **Launch Terminal** icon (![\[Dollar sign icon representing currency or financial transactions.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-launch-terminal.png)). This opens a terminal in the SageMaker image that the notebook runs in.
1. Run the following commands to display the contents of the `resource-metadata.json` file.
```
$ cd /opt/ml/metadata/
cat resource-metadata.json
```
The file should look similar to the following.
```
{
"AppType": "KernelGateway",
"DomainId": "d-xxxxxxxxxxxx",
"UserProfileName": "profile-name",
"ResourceArn": "arn:aws:sagemaker:us-east-2:account-id:app/d-xxxxxxxxxxxx/profile-name/KernelGateway/datascience--1-0-ml-t3-medium",
"ResourceName": "datascience--1-0-ml",
"AppImageVersion":""
}
```
# Get Notebook Differences in Amazon SageMaker Studio Classic
**Important**
Custom IAM policies that allow Amazon SageMaker Studio or Amazon SageMaker Studio Classic to create Amazon SageMaker resources must also grant permissions to add tags to those resources. The permission to add tags to resources is required because Studio and Studio Classic automatically tag any resources they create. If an IAM policy allows Studio and Studio Classic to create resources but does not allow tagging, "AccessDenied" errors can occur when trying to create resources. For more information, see [Provide permissions for tagging SageMaker AI resources](security_iam_id-based-policy-examples.md#grant-tagging-permissions).
[AWS managed policies for Amazon SageMaker AI](security-iam-awsmanpol.md) that give permissions to create SageMaker resources already include permissions to add tags while creating those resources.
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
You can display the difference between the current notebook and the last checkpoint or the last Git commit using the Amazon SageMaker AI UI.
The following screenshot shows the menu from a Studio Classic notebook.
![\[The location of the relevant menu in a Studio Classic notebook.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/studio-notebook-menu-diffs.png)
**Topics**
+ [Get the Difference Between the Last Checkpoint](#notebooks-diff-checkpoint)
+ [Get the Difference Between the Last Commit](#notebooks-diff-git)
## Get the Difference Between the Last Checkpoint
When you create a notebook, a hidden checkpoint file that matches the notebook is created. You can view changes between the notebook and the checkpoint file or revert the notebook to match the checkpoint file.
By default, a notebook is auto-saved every 120 seconds and also when you close the notebook. However, the checkpoint file isn't updated to match the notebook. To save the notebook and update the checkpoint file to match, you must choose the **Save notebook and create checkpoint** icon ( ![\[Padlock icon representing security or access control in cloud services.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-save-and-checkpoint.png)) on the left of the notebook menu or use the `Ctrl + S` keyboard shortcut.
To view the changes between the notebook and the checkpoint file, choose the **Checkpoint diff** icon (![\[Clock icon representing time or duration in a user interface.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-checkpoint-diff.png)) in the center of the notebook menu.
To revert the notebook to the checkpoint file, from the main Studio Classic menu, choose **File** then **Revert Notebook to Checkpoint**.
## Get the Difference Between the Last Commit
If a notebook is opened from a Git repository, you can view the difference between the notebook and the last Git commit.
To view the changes in the notebook from the last Git commit, choose the **Git diff** icon (![\[Dark button with white text displaying "git" in lowercase letters.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/notebook-git-diff.png)) in the center of the notebook menu.
# Manage Resources for Amazon SageMaker Studio Classic Notebooks
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
You can change the instance type, and SageMaker image and kernel from within an Amazon SageMaker Studio Classic notebook. To create a custom kernel to use with your notebooks, see [Custom Images in Amazon SageMaker Studio Classic](studio-byoi.md).
**Topics**
+ [Change the Instance Type for an Amazon SageMaker Studio Classic Notebook](notebooks-run-and-manage-switch-instance-type.md)
+ [Change the Image or a Kernel for an Amazon SageMaker Studio Classic Notebook](notebooks-run-and-manage-change-image.md)
+ [Shut Down Resources from Amazon SageMaker Studio Classic](notebooks-run-and-manage-shut-down.md)
# Change the Instance Type for an Amazon SageMaker Studio Classic Notebook
When you open a new Studio Classic notebook for the first time, you are assigned a default Amazon Elastic Compute Cloud (Amazon EC2) instance type to run the notebook. When you open additional notebooks on the same instance type, the notebooks run on the same instance as the first notebook, even if the notebooks use different kernels.
You can change the instance type that your Studio Classic notebook runs on from within the notebook.
The following information only applies to Studio Classic notebooks. For information about how to change the instance type of a Amazon SageMaker notebook instance, see [Update a Notebook Instance](nbi-update.md).
**Important**
If you change the instance type, unsaved information and existing settings for the notebook are lost, and installed packages must be re-installed.
The previous instance type continues to run even if no kernel sessions or apps are active. You must explicitly stop the instance to stop accruing charges. To stop the instance, see [Shut down resources](notebooks-run-and-manage-shut-down.md#notebooks-run-and-manage-shut-down-sessions).
The following screenshot shows the menu from a Studio Classic notebook. The processor and memory of the instance type powering the notebook are displayed as **2 vCPU \$1 4 GiB**.
![\[The location of the processor and memory of the instance type for the Studio Classic notebook.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/studio-notebook-menu-instance.png)
**To change the instance type**
1. Choose the processor and memory of the instance type powering the notebook. This opens a pop up window.
1. From the **Set up notebook environment** pop up window, select the **Instance type** dropdown menu.
1. From the **Instance type** dropdown, choose one of the instance types that are listed.
1. After choosing a type, choose **Select**.
1. Wait for the new instance to become enabled, and then the new instance type information is displayed.
For a list of the available instance types, see [Instance Types Available for Use With Amazon SageMaker Studio Classic Notebooks](notebooks-available-instance-types.md).
# Change the Image or a Kernel for an Amazon SageMaker Studio Classic Notebook
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
With Amazon SageMaker Studio Classic notebooks, you can change the notebook's image or kernel from within the notebook.
The following screenshot shows the menu from a Studio Classic notebook. The current SageMaker AI kernel and image are displayed as **Python 3 (Data Science)**, where `Python 3` denotes the kernel and `Data Science` denotes the SageMaker AI image that contains the kernel. The color of the circle to the right indicates the kernel is idle or busy. The kernel is busy when the center and the edge of the circle are the same color.
![\[The location of the current kernel and image in the menu bar from a Studio Classic notebook.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/studio-notebook-menu-kernel.png)
**To change a notebook's image or kernel**
1. Choose the image/kernel name in the notebook menu.
1. From the **Set up notebook environment** pop up window, select the **Image** or **Kernel** dropdown menu.
1. From the dropdown menu, choose one of the images or kernels that are listed.
1. After choosing an image or kernel, choose **Select**.
1. Wait for the kernel's status to show as idle, which indicates the kernel has started.
For a list of available SageMaker images and kernels, see [Amazon SageMaker Images Available for Use With Studio Classic Notebooks](notebooks-available-images.md).
# Shut Down Resources from Amazon SageMaker Studio Classic
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
You can shut down individual Amazon SageMaker AI resources, including notebooks, terminals, kernels, apps, and instances from Studio Classic. You can also shut down all of the resources in one of these categories at the same time. Amazon SageMaker Studio Classic does not support shutting down resources from within a notebook.
**Note**
When you shut down a Studio Classic notebook instance, additional resources that you created in Studio Classic are not deleted. For example, additional resources can include SageMaker AI endpoints, Amazon EMR clusters, and Amazon S3 buckets. To stop the accrual of charges, you must manually delete these resources. For information about finding resources that are accruing charges, see [Analyzing your costs with AWS Cost Explorer](https://docs.aws.amazon.com/cost-management/latest/userguide/ce-what-is.html).
The following topics demonstrate how to delete these SageMaker AI resources.
**Topics**
+ [Shut down an open notebook](#notebooks-run-and-manage-shut-down-notebook)
+ [Shut down resources](#notebooks-run-and-manage-shut-down-sessions)
## Shut down an open notebook
When you shut down a Studio Classic notebook, the notebook is not deleted. The kernel that the notebook is running on is shut down and any unsaved information in the notebook is lost. You can shut down an open notebook from the Studio Classic **File** menu or from the Running Terminal and Kernels pane. The following procedure shows how to shut down an open notebook from the Studio Classic **File** menu.
**To shut down an open notebook from the File menu**
1. Launch Studio Classic by following the steps in [Launch Amazon SageMaker Studio Classic](studio-launch.md).
1. (Optional) Save the notebook contents by choosing **File**, then **Save Notebook**.
1. Choose **File**.
1. Choose **Close and Shutdown Notebook**. This opens a pop-up window.
1. From the pop-up window, choose **OK**.
## Shut down resources
You can reach the **Running Terminals and Kernels** pane of Amazon SageMaker Studio Classic by selecting the **Running Terminals and Kernels** icon (![\[Black square icon representing a placeholder or empty image.\]](http://docs.aws.amazon.com/sagemaker/latest/dg/images/studio/icons/running-terminals-kernels.png)). The **Running Terminals and Kernels** pane consists of four sections. Each section lists all the resources of that type. You can shut down each resource individually or shut down all the resources in a section at the same time.
When you choose to shut down all resources in a section, the following occurs:
+ **RUNNING INSTANCES/RUNNING APPS** – All instances, apps, notebooks, kernel sessions, consoles/shells, and image terminals are shut down. System terminals aren't shut down.
+ **KERNEL SESSIONS** – All kernels, notebooks and consoles/shells are shut down.
+ **TERMINAL SESSIONS** – All image terminals and system terminals are shut down.
**To shut down resources**
1. Launch Studio Classic by following the steps in [Launch Amazon SageMaker Studio Classic](studio-launch.md).
1. Choose the **Running Terminals and Kernels** icon.
1. Do either of the following:
+ To shut down a specific resource, choose the **Shut Down** icon on the same row as the resource.
For running instances, a confirmation dialog box lists all of the resources that SageMaker AI will shut down. A confirmation dialog box displays all running apps. To proceed, choose **Shut down all**.
**Note**
A confirmation dialog box isn't displayed for kernel sessions or terminal sessions.
+ To shut down all resources in a section, choose the **X** to the right of the section label. A confirmation dialog box is displayed. Choose **Shut down all** to proceed.
**Note**
When you shut down these Studio Classic resources, any additional resources created from Studio Classic, such as SageMaker AI endpoints, Amazon EMR clusters, and Amazon S3 buckets are not deleted. You must manually delete these resources to stop the accrual of charges. For information about finding resources that are accruing charges, see [Analyzing your costs with AWS Cost Explorer](https://docs.aws.amazon.com/cost-management/latest/userguide/ce-what-is.html).
# Usage Metering for Amazon SageMaker Studio Classic Notebooks
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
There is no additional charge for using Amazon SageMaker Studio Classic. The costs incurred for running Amazon SageMaker Studio Classic notebooks, interactive shells, consoles, and terminals are based on Amazon Elastic Compute Cloud (Amazon EC2) instance usage.
When you run the following resources, you must choose a SageMaker image and kernel:
**From the Studio Classic Launcher**
+ Notebook
+ Interactive Shell
+ Image Terminal
**From the **File** menu**
+ Notebook
+ Console
When launched, the resource is run on an Amazon EC2 instance of the chosen instance type. If an instance of that type was previously launched and is available, the resource is run on that instance.
For CPU based images, the default suggested instance type is `ml.t3.medium`. For GPU based images, the default suggested instance type is `ml.g4dn.xlarge`.
The costs incurred are based on the instance type. You are billed separately for each instance.
Metering starts when an instance is created. Metering ends when all the apps on the instance are shut down, or the instance is shut down. For information about how to shut down an instance, see [Shut Down Resources from Amazon SageMaker Studio Classic](notebooks-run-and-manage-shut-down.md).
**Important**
You must shut down the instance to stop incurring charges. If you shut down the notebook running on the instance but don't shut down the instance, you will still incur charges. When you shut down the Studio Classic notebook instances, any additional resources, such as SageMaker AI endpoints, Amazon EMR clusters, and Amazon S3 buckets created from Studio Classic are not deleted. Delete those resources to stop accrual of charges.
When you open multiple notebooks on the same instance type, the notebooks run on the same instance even if they are using different kernels. You are billed only for the time that one instance is running.
You can change the instance type from within the notebook after you open it. For more information, see [Change the Instance Type for an Amazon SageMaker Studio Classic Notebook](notebooks-run-and-manage-switch-instance-type.md).
For information about billing along with pricing examples, see [Amazon SageMaker Pricing](https://aws.amazon.com/sagemaker/pricing/).
# Available Resources for Amazon SageMaker Studio Classic Notebooks
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
The following sections list the available resources for Amazon SageMaker Studio Classic notebooks.
**Topics**
+ [Instance Types Available for Use With Amazon SageMaker Studio Classic Notebooks](notebooks-available-instance-types.md)
+ [Amazon SageMaker Images Available for Use With Studio Classic Notebooks](notebooks-available-images.md)
# Instance Types Available for Use With Amazon SageMaker Studio Classic Notebooks
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
Amazon SageMaker Studio Classic notebooks run on Amazon Elastic Compute Cloud (Amazon EC2) instances. The following Amazon EC2 instance types are available for use with Studio Classic notebooks. For detailed information on which instance types fit your use case, and their performance capabilities, see [Amazon Elastic Compute Cloud Instance types](https://aws.amazon.com/ec2/instance-types/). For information about pricing for these instance types, see [Amazon EC2 Pricing](https://aws.amazon.com/ec2/pricing/).
For information about available Amazon SageMaker Notebook Instance types, see [CreateNotebookInstance](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateNotebookInstance.html#sagemaker-CreateNotebookInstance-request-InstanceType).
**Note**
For most use cases, you should use a `ml.t3.medium`. This is the default instance type for CPU-based SageMaker images, and is available as part of the [AWS Free Tier](https://aws.amazon.com/free).
**Topics**
+ [CPU instances](#notebooks-resources-no-gpu)
+ [Instances with 1 or more GPUs](#notebooks-resources-gpu)
## CPU instances
The following table lists the Amazon EC2 CPU instance types with no GPU attached that are available for use with Studio Classic notebooks. It also lists information about the specifications of each instance type. The default instance type for CPU-based images is `ml.t3.medium`.
For detailed information on which instance types fit your use case, and their performance capabilities, see [Amazon Elastic Compute Cloud Instance types](https://aws.amazon.com/ec2/instance-types/). For information about pricing for these instance types, see [Amazon EC2 Pricing](https://aws.amazon.com/ec2/pricing/).
CPU instances
| Instance | Use case | Fast launch | vCPU | Memory (GiB) | Instance Storage (GB) |
| --- | --- | --- | --- | --- | --- |
| ml.t3.medium | General purpose | Yes | 2 | 4 | Amazon EBS Only |
| ml.t3.large | General purpose | No | 2 | 8 | Amazon EBS Only |
| ml.t3.xlarge | General purpose | No | 4 | 16 | Amazon EBS Only |
| ml.t3.2xlarge | General purpose | No | 8 | 32 | Amazon EBS Only |
| ml.m5.large | General purpose | Yes | 2 | 8 | Amazon EBS Only |
| ml.m5.xlarge | General purpose | No | 4 | 16 | Amazon EBS Only |
| ml.m5.2xlarge | General purpose | No | 8 | 32 | Amazon EBS Only |
| ml.m5.4xlarge | General purpose | No | 16 | 64 | Amazon EBS Only |
| ml.m5.8xlarge | General purpose | No | 32 | 128 | Amazon EBS Only |
| ml.m5.12xlarge | General purpose | No | 48 | 192 | Amazon EBS Only |
| ml.m5.16xlarge | General purpose | No | 64 | 256 | Amazon EBS Only |
| ml.m5.24xlarge | General purpose | No | 96 | 384 | Amazon EBS Only |
| ml.m5d.large | General purpose | No | 2 | 8 | 1 x 75 NVMe SSD |
| ml.m5d.xlarge | General purpose | No | 4 | 16 | 1 x 150 NVMe SSD |
| ml.m5d.2xlarge | General purpose | No | 8 | 32 | 1 x 300 NVMe SSD |
| ml.m5d.4xlarge | General purpose | No | 16 | 64 | 2 x 300 NVMe SSD |
| ml.m5d.8xlarge | General purpose | No | 32 | 128 | 2 x 600 NVMe SSD |
| ml.m5d.12xlarge | General purpose | No | 48 | 192 | 2 x 900 NVMe SSD |
| ml.m5d.16xlarge | General purpose | No | 64 | 256 | 4 x 600 NVMe SSD |
| ml.m5d.24xlarge | General purpose | No | 96 | 384 | 4 x 900 NVMe SSD |
| ml.c5.large | Compute optimized | Yes | 2 | 4 | Amazon EBS Only |
| ml.c5.xlarge | Compute optimized | No | 4 | 8 | Amazon EBS Only |
| ml.c5.2xlarge | Compute optimized | No | 8 | 16 | Amazon EBS Only |
| ml.c5.4xlarge | Compute optimized | No | 16 | 32 | Amazon EBS Only |
| ml.c5.9xlarge | Compute optimized | No | 36 | 72 | Amazon EBS Only |
| ml.c5.12xlarge | Compute optimized | No | 48 | 96 | Amazon EBS Only |
| ml.c5.18xlarge | Compute optimized | No | 72 | 144 | Amazon EBS Only |
| ml.c5.24xlarge | Compute optimized | No | 96 | 192 | Amazon EBS Only |
| ml.r5.large | Memory optimized | No | 2 | 16 | Amazon EBS Only |
| ml.r5.xlarge | Memory optimized | No | 4 | 32 | Amazon EBS Only |
| ml.r5.2xlarge | Memory optimized | No | 8 | 64 | Amazon EBS Only |
| ml.r5.4xlarge | Memory optimized | No | 16 | 128 | Amazon EBS Only |
| ml.r5.8xlarge | Memory optimized | No | 32 | 256 | Amazon EBS Only |
| ml.r5.12xlarge | Memory optimized | No | 48 | 384 | Amazon EBS Only |
| ml.r5.16xlarge | Memory optimized | No | 64 | 512 | Amazon EBS Only |
| ml.r5.24xlarge | Memory optimized | No | 96 | 768 | Amazon EBS Only |
## Instances with 1 or more GPUs
The following table lists the Amazon EC2 instance types with 1 or more GPUs attached that are available for use with Studio Classic notebooks. It also lists information about the specifications of each instance type. The default instance type for GPU-based images is `ml.g4dn.xlarge`.
For detailed information on which instance types fit your use case, and their performance capabilities, see [Amazon Elastic Compute Cloud Instance types](https://aws.amazon.com/ec2/instance-types/). For information about pricing for these instance types, see [Amazon EC2 Pricing](https://aws.amazon.com/ec2/pricing/).
Instances with 1 or more GPUs
| Instance | Use case | Fast launch | GPUs | vCPU | Memory (GiB) | GPU Memory (GiB) | Instance Storage (GB) |
| --- | --- | --- | --- | --- | --- | --- | --- |
| ml.p3.2xlarge | Accelerated computing | No | 1 | 8 | 61 | 16 | Amazon EBS Only |
| ml.p3.8xlarge | Accelerated computing | No | 4 | 32 | 244 | 64 | Amazon EBS Only |
| ml.p3.16xlarge | Accelerated computing | No | 8 | 64 | 488 | 128 | Amazon EBS Only |
| ml.p3dn.24xlarge | Accelerated computing | No | 8 | 96 | 768 | 256 | 2 x 900 NVMe SSD |
| ml.p4d.24xlarge | Accelerated computing | No | 8 | 96 | 1152 | 320 GB HBM2 | 8 x 1000 NVMe SSD |
| ml.p4de.24xlarge | Accelerated computing | No | 8 | 96 | 1152 | 640 GB HBM2e | 8 x 1000 NVMe SSD |
| ml.g4dn.xlarge | Accelerated computing | Yes | 1 | 4 | 16 | 16 | 1 x 125 NVMe SSD |
| ml.g4dn.2xlarge | Accelerated computing | No | 1 | 8 | 32 | 16 | 1 x 225 NVMe SSD |
| ml.g4dn.4xlarge | Accelerated computing | No | 1 | 16 | 64 | 16 | 1 x 225 NVMe SSD |
| ml.g4dn.8xlarge | Accelerated computing | No | 1 | 32 | 128 | 16 | 1 x 900 NVMe SSD |
| ml.g4dn.12xlarge | Accelerated computing | No | 4 | 48 | 192 | 64 | 1 x 900 NVMe SSD |
| ml.g4dn.16xlarge | Accelerated computing | No | 1 | 64 | 256 | 16 | 1 x 900 NVMe SSD |
| ml.g5.xlarge | Accelerated computing | No | 1 | 4 | 16 | 24 | 1 x 250 NVMe SSD |
| ml.g5.2xlarge | Accelerated computing | No | 1 | 8 | 32 | 24 | 1 x 450 NVMe SSD |
| ml.g5.4xlarge | Accelerated computing | No | 1 | 16 | 64 | 24 | 1 x 600 NVMe SSD |
| ml.g5.8xlarge | Accelerated computing | No | 1 | 32 | 128 | 24 | 1 x 900 NVMe SSD |
| ml.g5.12xlarge | Accelerated computing | No | 4 | 48 | 192 | 96 | 1 x 3800 NVMe SSD |
| ml.g5.16xlarge | Accelerated computing | No | 1 | 64 | 256 | 24 | 1 x 1900 NVMe SSD |
| ml.g5.24xlarge | Accelerated computing | No | 4 | 96 | 384 | 96 | 1 x 3800 NVMe SSD |
| ml.g5.48xlarge | Accelerated computing | No | 8 | 192 | 768 | 192 | 2 x 3800 NVMe SSD |
# Amazon SageMaker Images Available for Use With Studio Classic Notebooks
**Important**
As of November 30, 2023, the previous Amazon SageMaker Studio experience is now named Amazon SageMaker Studio Classic. The following section is specific to using the Studio Classic application. For information about using the updated Studio experience, see [Amazon SageMaker Studio](studio-updated.md).
Studio Classic is still maintained for existing workloads but is no longer available for onboarding. You can only stop or delete existing Studio Classic applications and cannot create new ones. We recommend that you [migrate your workload to the new Studio experience](studio-updated-migrate.md).
This page lists the SageMaker images and associated kernels that are available in Amazon SageMaker Studio Classic. This page also gives information about the format needed to create the ARN for each image. SageMaker images contain the latest [Amazon SageMaker Python SDK](https://sagemaker.readthedocs.io/en/stable) and the latest version of the kernel. For more information, see [Deep Learning Containers Images](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/deep-learning-containers-images.html).
**Topics**
+ [Image ARN format](#notebooks-available-images-arn)
+ [Supported URI tags](#notebooks-available-uri-tag)
+ [Supported images](#notebooks-available-images-supported)
+ [Images slated for deprecation](#notebooks-available-images-deprecation)
+ [Deprecated images](#notebooks-available-images-deprecated)
## Image ARN format
The following table lists the image ARN and URI format for each Region. To create the full ARN for an image, replace the *resource-identifier* placeholder with the corresponding resource identifier for the image. The resource identifier is found in the SageMaker images and kernels table. To create the full URI for an image, replace the *tag* placeholder with the corresponding cpu or gpu tag. For the list of tags you can use, see [Supported URI tags](#notebooks-available-uri-tag).
**Note**
SageMaker Distribution images use a distinct set of image ARNs, which are listed in the following table.
| Region | Image ARN Format | SageMaker Distribution Image ARN Format | SageMaker Distribution Image URI Format |
| --- | --- | --- | --- |
| us-east-1 | arn:aws:sagemaker:us-east-1:081325390199:image/resource-identifier | arn:aws:sagemaker:us-east-1:885854791233:image/resource-identifier | 885854791233.dkr.ecr.us-east-1.amazonaws.com/sagemaker-distribution-prod:tag |
| us-east-2 | arn:aws:sagemaker:us-east-2:429704687514:image/resource-identifier | arn:aws:sagemaker:us-east-2:137914896644:image/resource-identifier | 137914896644.dkr.ecr.us-east-2.amazonaws.com/sagemaker-distribution-prod:tag |
| us-west-1 | arn:aws:sagemaker:us-west-1:742091327244:image/resource-identifier | arn:aws:sagemaker:us-west-1:053634841547:image/resource-identifier | 053634841547.dkr.ecr.us-west-1.amazonaws.com/sagemaker-distribution-prod:tag |
| us-west-2 | arn:aws:sagemaker:us-west-2:236514542706:image/resource-identifier | arn:aws:sagemaker:us-west-2:542918446943:image/resource-identifier | 542918446943.dkr.ecr.us-west-2.amazonaws.com/sagemaker-distribution-prod:tag |
| af-south-1 | arn:aws:sagemaker:af-south-1:559312083959:image/resource-identifier | arn:aws:sagemaker:af-south-1:238384257742:image/resource-identifier | 238384257742.dkr.ecr.af-south-1.amazonaws.com/sagemaker-distribution-prod:tag |
| ap-east-1 | arn:aws:sagemaker:ap-east-1:493642496378:image/resource-identifier | arn:aws:sagemaker:ap-east-1:523751269255:image/resource-identifier | 523751269255.dkr.ecr.ap-east-1.amazonaws.com/sagemaker-distribution-prod:tag |
| ap-south-1 | arn:aws:sagemaker:ap-south-1:394103062818:image/resource-identifier | arn:aws:sagemaker:ap-south-1:245090515133:image/resource-identifier | 245090515133.dkr.ecr.ap-south-1.amazonaws.com/sagemaker-distribution-prod:tag |
| ap-northeast-2 | arn:aws:sagemaker:ap-northeast-2:806072073708:image/resource-identifier | arn:aws:sagemaker:ap-northeast-2:064688005998:image/resource-identifier | 064688005998.dkr.ecr.ap-northeast-2.amazonaws.com/sagemaker-distribution-prod:tag |
| ap-southeast-1 | arn:aws:sagemaker:ap-southeast-1:492261229750:image/resource-identifier | arn:aws:sagemaker:ap-southeast-1:022667117163:image/resource-identifier | 022667117163.dkr.ecr.ap-southeast-1.amazonaws.com/sagemaker-distribution-prod:tag |
| ap-southeast-2 | arn:aws:sagemaker:ap-southeast-2:452832661640:image/resource-identifier | arn:aws:sagemaker:ap-southeast-2:648430277019:image/resource-identifier | 648430277019.dkr.ecr.ap-southeast-2.amazonaws.com/sagemaker-distribution-prod:tag |
| ap-northeast-1 | arn:aws:sagemaker:ap-northeast-1:102112518831:image/resource-identifier | arn:aws:sagemaker:ap-northeast-1:010972774902:image/resource-identifier | 010972774902.dkr.ecr.ap-northeast-1.amazonaws.com/sagemaker-distribution-prod:tag |
| ca-central-1 | arn:aws:sagemaker:ca-central-1:310906938811:image/resource-identifier | arn:aws:sagemaker:ca-central-1:481561238223:image/resource-identifier | 481561238223.dkr.ecr.ca-central-1.amazonaws.com/sagemaker-distribution-prod:tag |
| eu-central-1 | arn:aws:sagemaker:eu-central-1:936697816551:image/resource-identifier | arn:aws:sagemaker:eu-central-1:545423591354:image/resource-identifier | 545423591354.dkr.ecr.eu-central-1.amazonaws.com/sagemaker-distribution-prod:tag |
| eu-west-1 | arn:aws:sagemaker:eu-west-1:470317259841:image/resource-identifier | arn:aws:sagemaker:eu-west-1:819792524951:image/resource-identifier | 819792524951.dkr.ecr.eu-west-1.amazonaws.com/sagemaker-distribution-prod:tag |
| eu-west-2 | arn:aws:sagemaker:eu-west-2:712779665605:image/resource-identifier | arn:aws:sagemaker:eu-west-2:021081402939:image/resource-identifier | 021081402939.dkr.ecr.eu-west-2.amazonaws.com/sagemaker-distribution-prod:tag |
| eu-west-3 | arn:aws:sagemaker:eu-west-3:615547856133:image/resource-identifier | arn:aws:sagemaker:eu-west-3:856416204555:image/resource-identifier | 856416204555.dkr.ecr.eu-west-3.amazonaws.com/sagemaker-distribution-prod:tag |
| eu-north-1 | arn:aws:sagemaker:eu-north-1:243637512696:image/resource-identifier | arn:aws:sagemaker:eu-north-1:175620155138:image/resource-identifier | 175620155138.dkr.ecr.eu-north-1.amazonaws.com/sagemaker-distribution-prod:tag |
| eu-south-1 | arn:aws:sagemaker:eu-south-1:592751261982:image/resource-identifier | arn:aws:sagemaker:eu-south-1:810671768855:image/resource-identifier | 810671768855.dkr.ecr.eu-south-1.amazonaws.com/sagemaker-distribution-prod:tag |
| sa-east-1 | arn:aws:sagemaker:sa-east-1:782484402741:image/resource-identifier | arn:aws:sagemaker:sa-east-1:567556641782:image/resource-identifier | 567556641782.dkr.ecr.sa-east-1.amazonaws.com/sagemaker-distribution-prod:tag |
| ap-northeast-3 | arn:aws:sagemaker:ap-northeast-3:792733760839:image/resource-identifier | arn:aws:sagemaker:ap-northeast-3:564864627153:image/resource-identifier | 564864627153.dkr.ecr.ap-northeast-3.amazonaws.com/sagemaker-distribution-prod:tag |
| ap-southeast-3 | arn:aws:sagemaker:ap-southeast-3:276181064229:image/resource-identifier | arn:aws:sagemaker:ap-southeast-3:370607712162:image/resource-identifier | 370607712162.dkr.ecr.ap-southeast-3.amazonaws.com/sagemaker-distribution-prod:tag |
| me-south-1 | arn:aws:sagemaker:me-south-1:117516905037:image/resource-identifier | arn:aws:sagemaker:me-south-1:523774347010:image/resource-identifier | 523774347010.dkr.ecr.me-south-1.amazonaws.com/sagemaker-distribution-prod:tag |
| me-central-1 | arn:aws:sagemaker:me-central-1:103105715889:image/resource-identifier | arn:aws:sagemaker:me-central-1:358593528301:image/resource-identifier | 358593528301.dkr.ecr.me-central-1.amazonaws.com/sagemaker-distribution-prod:tag |
## Supported URI tags
The following list shows the tags you can include in your image URI.
+ 1-cpu
+ 1-gpu
+ 0-cpu
+ 0-gpu
**The following examples show URIs with various tag formats:**
+ 542918446943.dkr.ecr.us-west-2.amazonaws.com/sagemaker-distribution-prod:1-cpu
+ 542918446943.dkr.ecr.us-west-2.amazonaws.com/sagemaker-distribution-prod:0-gpu
## Supported images
The following table gives information about the SageMaker images and associated kernels that are available in Amazon SageMaker Studio Classic. It also gives information about the resource identifier and Python version included in the image.
SageMaker images and kernels
| SageMaker Image | Description | Resource Identifier | Kernels (and Identifier) | Python Version |
| --- | --- | --- | --- | --- |
| Base Python 4.3 | Official Python 3.11 image from DockerHub with boto3 and AWS CLI included. | sagemaker-base-python-v4 | Python 3 (python3) | Python 3.11 |
| Base Python 4.2 | Official Python 3.11 image from DockerHub with boto3 and AWS CLI included. | sagemaker-base-python-v4 | Python 3 (python3) | Python 3.11 |
| Base Python 4.1 | Official Python 3.11 image from DockerHub with boto3 and AWS CLI included. | sagemaker-base-python-v4 | Python 3 (python3) | Python 3.11 |
| Base Python 4.0 | Official Python 3.11 image from DockerHub with boto3 and AWS CLI included. | sagemaker-base-python-v4 | Python 3 (python3) | Python 3.11 |
| Base Python 3.0 | Official Python 3.10 image from DockerHub with boto3 and AWS CLI included. | sagemaker-base-python-310-v1 | Python 3 (python3) | Python 3.10 |
| Data Science 5.3 | Data Science 5.3 is a Python 3.11 [conda](https://docs.conda.io/projects/conda/en/latest/index.html) image based on Ubuntu version jammy-20240212. It includes the most commonly used Python packages and libraries, such as NumPy and SciKit Learn. | sagemaker-data-science-v5 | Python 3 (python3) | Python 3.11 |
| Data Science 5.2 | Data Science 5.2 is a Python 3.11 [conda](https://docs.conda.io/projects/conda/en/latest/index.html) image based on Ubuntu version jammy-20240212. It includes the most commonly used Python packages and libraries, such as NumPy and SciKit Learn. | sagemaker-data-science-v5 | Python 3 (python3) | Python 3.11 |
| Data Science 5.1 | Data Science 5.1 is a Python 3.11 [conda](https://docs.conda.io/projects/conda/en/latest/index.html) image based on Ubuntu version jammy-20240212. It includes the most commonly used Python packages and libraries, such as NumPy and SciKit Learn. | sagemaker-data-science-v5 | Python 3 (python3) | Python 3.11 |
| Data Science 5.0 | Data Science 5.0 is a Python 3.11 [conda](https://docs.conda.io/projects/conda/en/latest/index.html) image based on Ubuntu version jammy-20240212. It includes the most commonly used Python packages and libraries, such as NumPy and SciKit Learn. | sagemaker-data-science-v5 | Python 3 (python3) | Python 3.11 |
| Data Science 4.0 | Data Science 4.0 is a Python 3.11 [conda](https://docs.conda.io/projects/conda/en/latest/index.html) image based on Ubuntu version 22.04. It includes the most commonly used Python packages and libraries, such as NumPy and SciKit Learn. | sagemaker-data-science-311-v1 | Python 3 (python3) | Python 3.11 |
| Data Science 3.0 | Data Science 3.0 is a Python 3.10 [conda](https://docs.conda.io/projects/conda/en/latest/index.html) image based on Ubuntu version 22.04. It includes the most commonly used Python packages and libraries, such as NumPy and SciKit Learn. | sagemaker-data-science-310-v1 | Python 3 (python3) | Python 3.10 |
| Geospatial 1.0 | Amazon SageMaker geospatial is a Python image consisting of commonly used geospatial libraries such as GDAL, Fiona, GeoPandas, Shapley, and Rasterio. It allows you to visualize geospatial data within SageMaker AI. For more information, see [Amazon SageMaker geospatial Notebook SDK](https://docs.aws.amazon.com/sagemaker/latest/dg/geospatial-notebook-sdk.html) | sagemaker-geospatial-1.0 | Python 3 (python3) | Python 3.10 |
| SparkAnalytics 4.3 | The SparkAnalytics 4.3 image provides Spark and PySpark kernel options on Amazon SageMaker Studio Classic, including SparkMagic Spark, SparkMagic PySpark, Glue Spark, and Glue PySpark, enabling flexible distributed data processing. | sagemaker-spark-analytics-v4 | [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/sagemaker/latest/dg/notebooks-available-images.html) | Python 3.11 |
| SparkAnalytics 4.2 | The SparkAnalytics 4.2 image provides Spark and PySpark kernel options on Amazon SageMaker Studio Classic, including SparkMagic Spark, SparkMagic PySpark, Glue Spark, and Glue PySpark, enabling flexible distributed data processing. | sagemaker-spark-analytics-v4 | [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/sagemaker/latest/dg/notebooks-available-images.html) | Python 3.11 |
| SparkAnalytics 4.1 | The SparkAnalytics 4.1 image provides Spark and PySpark kernel options on Amazon SageMaker Studio Classic, including SparkMagic Spark, SparkMagic PySpark, Glue Spark, and Glue PySpark, enabling flexible distributed data processing. | sagemaker-spark-analytics-v4 | [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/sagemaker/latest/dg/notebooks-available-images.html) | Python 3.11 |
| SparkAnalytics 4.0 | The SparkAnalytics 4.0 image provides Spark and PySpark kernel options on Amazon SageMaker Studio Classic, including SparkMagic Spark, SparkMagic PySpark, Glue Spark, and Glue PySpark, enabling flexible distributed data processing. | sagemaker-spark-analytics-v4 | [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/sagemaker/latest/dg/notebooks-available-images.html) | Python 3.11 |
| SparkAnalytics 3.0 | The SparkAnalytics 3.0 image provides Spark and PySpark kernel options on Amazon SageMaker Studio Classic, including SparkMagic Spark, SparkMagic PySpark, Glue Spark, and Glue PySpark, enabling flexible distributed data processing. | sagemaker-sparkanalytics-311-v1 | [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/sagemaker/latest/dg/notebooks-available-images.html) | Python 3.11 |
| SparkAnalytics 2.0 | Anaconda Individual Edition with PySpark and Spark kernels. For more information, see [sparkmagic](https://github.com/jupyter-incubator/sparkmagic). | sagemaker-sparkanalytics-310-v1 | [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/sagemaker/latest/dg/notebooks-available-images.html) | Python 3.10 |
| PyTorch 2.4.0 Python 3.11 CPU Optimized | The AWS Deep Learning Containers for PyTorch 2.4.0 with CUDA 12.4 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.4.0-cpu-py311 | Python 3 (python3) | Python 3.11 |
| PyTorch 2.4.0 Python 3.11 GPU Optimized | The AWS Deep Learning Containers for PyTorch 2.4.0 with CUDA 12.4 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.4.0-gpu-py311 | Python 3 (python3) | Python 3.11 |
| PyTorch 2.3.0 Python 3.11 CPU Optimized | The AWS Deep Learning Containers for PyTorch 2.3.0 with CUDA 12.1 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.3.0-cpu-py311 | Python 3 (python3) | Python 3.11 |
| PyTorch 2.3.0 Python 3.11 GPU Optimized | The AWS Deep Learning Containers for PyTorch 2.3.0 with CUDA 12.1 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.3.0-gpu-py311 | Python 3 (python3) | Python 3.11 |
| PyTorch 2.2.0 Python 3.10 CPU Optimized | The AWS Deep Learning Containers for PyTorch 2.2 with CUDA 12.1 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.2.0-cpu-py310 | Python 3 (python3) | Python 3.10 |
| PyTorch 2.2.0 Python 3.10 GPU Optimized | The AWS Deep Learning Containers for PyTorch 2.2 with CUDA 12.1 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.2.0-gpu-py310 | Python 3 (python3) | Python 3.10 |
| PyTorch 2.1.0 Python 3.10 CPU Optimized | The AWS Deep Learning Containers for PyTorch 2.1 with CUDA 12.1 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.1.0-cpu-py310 | Python 3 (python3) | Python 3.10 |
| PyTorch 2.1.0 Python 3.10 GPU Optimized | The AWS Deep Learning Containers for PyTorch 2.1 with CUDA 12.1 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.1.0-gpu-py310 | Python 3 (python3) | Python 3.10 |
| PyTorch 1.13 HuggingFace Python 3.10 Neuron Optimized | PyTorch 1.13 image with HuggingFace and Neuron packages installed for training on Trainium instances optimized for performance and scale on AWS. | pytorch-1.13-hf-neuron-py310 | Python 3 (python3) | Python 3.10 |
| PyTorch 1.13 Python 3.10 Neuron Optimized | PyTorch 1.13 image with Neuron packages installed for training on Trainium instances optimized for performance and scale on AWS. | pytorch-1.13-neuron-py310 | Python 3 (python3) | Python 3.10 |
| TensorFlow 2.14.0 Python 3.10 CPU Optimized | The AWS Deep Learning Containers for TensorFlow 2.14 with CUDA 11.8 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | tensorflow-2.14.1-cpu-py310-ubuntu20.04-sagemaker-v1.0 | Python 3 (python3) | Python 3.10 |
| TensorFlow 2.14.0 Python 3.10 GPU Optimized | The AWS Deep Learning Containers for TensorFlow 2.14 with CUDA 11.8 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | tensorflow-2.14.1-gpu-py310-cu118-ubuntu20.04-sagemaker-v1.0 | Python 3 (python3) | Python 3.10 |
## Images slated for deprecation
SageMaker AI ends support for images the day after any of the packages in the image reach end-of life by their publisher. The following SageMaker images are slated for deprecation.
Images based on Python 3.8 reached [end-of-life](https://endoflife.date/python) on October 31st, 2024. Starting on November 1, 2024, SageMaker AI will discontinue support for these images and they will not be selectable from the Studio Classic UI. To avoid non-compliance issues, if you're using any of these images, we recommend that you move to an image with a later version.
SageMaker images slated for deprecation
| SageMaker Image | Deprecation date | Description | Resource Identifier | Kernels | Python Version |
| --- | --- | --- | --- | --- | --- |
| SageMaker Distribution v0.12 CPU | November 1, 2024 | SageMaker Distribution v0 CPU is a Python 3.8 image that includes popular frameworks for machine learning, data science and visualization on CPU. This includes deep learning frameworks like PyTorch, TensorFlow and Keras; popular Python packages like numpy, scikit-learn and pandas; and IDEs like Jupyter Lab. For more information, see the [Amazon SageMaker AI Distribution](https://github.com/aws/sagemaker-distribution) repo. | sagemaker-distribution-cpu-v0 | Python 3 (python3) | Python 3.8 |
| SageMaker Distribution v0.12 GPU | November 1, 2024 | SageMaker Distribution v0 GPU is a Python 3.8 image that includes popular frameworks for machine learning, data science and visualization on GPU. This includes deep learning frameworks like PyTorch, TensorFlow and Keras; popular Python packages like numpy, scikit-learn and pandas; and IDEs like Jupyter Lab. For more information, see the [Amazon SageMaker AI Distribution](https://github.com/aws/sagemaker-distribution) repo. | sagemaker-distribution-gpu-v0 | Python 3 (python3) | Python 3.8 |
| Base Python 2.0 | November 1, 2024 | Official Python 3.8 image from DockerHub with boto3 and AWS CLI included. | sagemaker-base-python-38 | Python 3 (python3) | Python 3.8 |
| Data Science 2.0 | November 1, 2024 | Data Science 2.0 is a Python 3.8 [conda](https://docs.conda.io/projects/conda/en/latest/index.html) image based on Ubuntu version 22.04. It includes the most commonly used Python packages and libraries, such as NumPy and SciKit Learn. | sagemaker-data-science-38 | Python 3 (python3) | Python 3.8 |
| PyTorch 1.13 Python 3.9 CPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for PyTorch 1.13 with CUDA 11.3 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-1.13-cpu-py39 | Python 3 (python3) | Python 3.9 |
| PyTorch 1.13 Python 3.9 GPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for PyTorch 1.13 with CUDA 11.7 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-1.13-gpu-py39 | Python 3 (python3) | Python 3.9 |
| PyTorch 1.12 Python 3.8 CPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for PyTorch 1.12 with CUDA 11.3 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [AWS Deep Learning Containers for PyTorch 1.12.0](https://aws.amazon.com/releasenotes/aws-deep-learning-containers-for-pytorch-1-12-0-on-sagemaker/). | pytorch-1.12-cpu-py38 | Python 3 (python3) | Python 3.8 |
| PyTorch 1.12 Python 3.8 GPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for PyTorch 1.12 with CUDA 11.3 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [AWS Deep Learning Containers for PyTorch 1.12.0](https://aws.amazon.com/releasenotes/aws-deep-learning-containers-for-pytorch-1-12-0-on-sagemaker/). | pytorch-1.12-gpu-py38 | Python 3 (python3) | Python 3.8 |
| PyTorch 1.10 Python 3.8 CPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for PyTorch 1.10 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [AWS Deep Learning Containers for PyTorch 1.10.2 on SageMaker AI](https://aws.amazon.com/releasenotes/aws-deep-learning-containers-for-pytorch-1-10-2-on-sagemaker/). | pytorch-1.10-cpu-py38 | Python 3 (python3) | Python 3.8 |
| PyTorch 1.10 Python 3.8 GPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for PyTorch 1.10 with CUDA 11.3 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [AWS Deep Learning Containers for PyTorch 1.10.2 on SageMaker AI](https://aws.amazon.com/releasenotes/aws-deep-learning-containers-for-pytorch-1-10-2-on-sagemaker/). | pytorch-1.10-gpu-py38 | Python 3 (python3) | Python 3.8 |
| SparkAnalytics 1.0 | November 1, 2024 | Anaconda Individual Edition with PySpark and Spark kernels. For more information, see [sparkmagic](https://github.com/jupyter-incubator/sparkmagic). | sagemaker-sparkanalytics-v1 | [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/sagemaker/latest/dg/notebooks-available-images.html) | Python 3.8 |
| TensorFlow 2.13.0 Python 3.10 CPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for TensorFlow 2.13 with CUDA 11.8 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers.](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | tensorflow-2.13.0-cpu-py310-ubuntu20.04-sagemaker-v1.0 | Python 3 (python3) | Python 3.10 |
| TensorFlow 2.13.0 Python 3.10 GPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for TensorFlow 2.13 with CUDA 11.8 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers.](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html) | tensorflow-2.13.0-gpu-py310-cu118-ubuntu20.04-sagemaker-v1.0 | Python 3 (python3) | Python 3.10 |
| TensorFlow 2.6 Python 3.8 CPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for TensorFlow 2.6 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [AWS Deep Learning Containers for TensorFlow 2.6](https://aws.amazon.com/releasenotes/aws-deep-learning-containers-for-tensorflow-2-6/). | tensorflow-2.6-cpu-py38-ubuntu20.04-v1 | Python 3 (python3) | Python 3.8 |
| TensorFlow 2.6 Python 3.8 GPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for TensorFlow 2.6 with CUDA 11.2 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [AWS Deep Learning Containers for TensorFlow 2.6](https://aws.amazon.com/releasenotes/aws-deep-learning-containers-for-tensorflow-2-6/). | tensorflow-2.6-gpu-py38-cu112-ubuntu20.04-v1 | Python 3 (python3) | Python 3.8 |
| PyTorch 2.0.1 Python 3.10 CPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for PyTorch 2.0.1 with CUDA 12.1 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.0.1-cpu-py310 | Python 3 (python3) | Python 3.10 |
| PyTorch 2.0.1 Python 3.10 GPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for PyTorch 2.0.1 with CUDA 12.1 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.0.1-gpu-py310 | Python 3 (python3) | Python 3.10 |
| PyTorch 2.0.0 Python 3.10 CPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for PyTorch 2.0.0 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.0.0-cpu-py310 | Python 3 (python3) | Python 3.10 |
| PyTorch 2.0.0 Python 3.10 GPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for PyTorch 2.0.0 with CUDA 11.8 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | pytorch-2.0.0-gpu-py310 | Python 3 (python3) | Python 3.10 |
| TensorFlow 2.12.0 Python 3.10 CPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for TensorFlow 2.12.0 with CUDA 11.2 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | tensorflow-2.12.0-cpu-py310-ubuntu20.04-sagemaker-v1.0 | Python 3 (python3) | Python 3.10 |
| TensorFlow 2.12.0 Python 3.10 GPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for TensorFlow 2.12.0 with CUDA 11.8 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | tensorflow-2.12.0-gpu-py310-cu118-ubuntu20.04-sagemaker-v1 | Python 3 (python3) | Python 3.10 |
| TensorFlow 2.11.0 Python 3.9 CPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for TensorFlow 2.11.0 with CUDA 11.2 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | tensorflow-2.11.0-cpu-py39-ubuntu20.04-sagemaker-v1.1 | Python 3 (python3) | Python 3.9 |
| TensorFlow 2.11.0 Python 3.9 GPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for TensorFlow 2.11.0 with CUDA 11.2 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | tensorflow-2.11.0-gpu-py39-cu112-ubuntu20.04-sagemaker-v1.1 | Python 3 (python3) | Python 3.9 |
| TensorFlow 2.10 Python 3.9 CPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for TensorFlow 2.10 with CUDA 11.2 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | tensorflow-2.10.1-cpu-py39-ubuntu20.04-sagemaker-v1.2 | Python 3 (python3) | Python 3.9 |
| TensorFlow 2.10 Python 3.9 GPU Optimized | November 1, 2024 | The AWS Deep Learning Containers for TensorFlow 2.10 with CUDA 11.2 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [Release Notes for Deep Learning Containers](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/dlc-release-notes.html). | tensorflow-2.10.1-gpu-py39-ubuntu20.04-sagemaker-v1.2 | Python 3 (python3) | Python 3.9 |
## Deprecated images
SageMaker AI has ended support for the following images. Deprecation occurs the day after any of the packages in the image reach end-of life by their publisher.
SageMaker images slated for deprecation
| SageMaker Image | Deprecation date | Description | Resource Identifier | Kernels | Python Version |
| --- | --- | --- | --- | --- | --- |
| Data Science | October 30, 2023 | Data Science is a Python 3.7 [conda](https://docs.conda.io/projects/conda/en/latest/index.html) image with the most commonly used Python packages and libraries, such as NumPy and SciKit Learn. | datascience-1.0 | Python 3 | Python 3.7 |
| SageMaker JumpStart Data Science 1.0 | October 30, 2023 | SageMaker JumpStart Data Science 1.0 is a JumpStart image that includes commonly used packages and libraries. | sagemaker-jumpstart-data-science-1.0 | Python 3 | Python 3.7 |
| SageMaker JumpStart MXNet 1.0 | October 30, 2023 | SageMaker JumpStart MXNet 1.0 is a JumpStart image that includes MXNet. | sagemaker-jumpstart-mxnet-1.0 | Python 3 | Python 3.7 |
| SageMaker JumpStart PyTorch 1.0 | October 30, 2023 | SageMaker JumpStart PyTorch 1.0 is a JumpStart image that includes PyTorch. | sagemaker-jumpstart-pytorch-1.0 | Python 3 | Python 3.7 |
| SageMaker JumpStart TensorFlow 1.0 | October 30, 2023 | SageMaker JumpStart TensorFlow 1.0 is a JumpStart image that includes TensorFlow. | sagemaker-jumpstart-tensorflow-1.0 | Python 3 | Python 3.7 |
| SparkMagic | October 30, 2023 | Anaconda Individual Edition with PySpark and Spark kernels. For more information, see [sparkmagic](https://github.com/jupyter-incubator/sparkmagic). | sagemaker-sparkmagic | [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/sagemaker/latest/dg/notebooks-available-images.html) | Python 3.7 |
| TensorFlow 2.3 Python 3.7 CPU Optimized | October 30, 2023 | The AWS Deep Learning Containers for TensorFlow 2.3 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [AWS Deep Learning Containers with TensorFlow 2.3.0](https://aws.amazon.com/releasenotes/aws-deep-learning-containers-with-tensorflow-2-3-0/). | tensorflow-2.3-cpu-py37-ubuntu18.04-v1 | Python 3 | Python 3.7 |
| TensorFlow 2.3 Python 3.7 GPU Optimized | October 30, 2023 | The AWS Deep Learning Containers for TensorFlow 2.3 with CUDA 11.0 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [AWS Deep Learning Containers for TensorFlow 2.3.1 with CUDA 11.0](https://aws.amazon.com/releasenotes/aws-deep-learning-containers-for-tensorflow-2-3-1-with-cuda-11-0/). | tensorflow-2.3-gpu-py37-cu110-ubuntu18.04-v3 | Python 3 | Python 3.7 |
| TensorFlow 1.15 Python 3.7 CPU Optimized | October 30, 2023 | The AWS Deep Learning Containers for TensorFlow 1.15 include containers for training on CPU, optimized for performance and scale on AWS. For more information, see [AWS Deep Learning Containers v7.0 for TensorFlow](https://aws.amazon.com/releasenotes/aws-deep-learning-containers-v7-0-for-tensorflow/). | tensorflow-1.15-cpu-py37-ubuntu18.04-v7 | Python 3 | Python 3.7 |
| TensorFlow 1.15 Python 3.7 GPU Optimized | October 30, 2023 | The AWS Deep Learning Containers for TensorFlow 1.15 with CUDA 11.0 include containers for training on GPU, optimized for performance and scale on AWS. For more information, see [AWS Deep Learning Containers v7.0 for TensorFlow](https://aws.amazon.com/releasenotes/aws-deep-learning-containers-v7-0-for-tensorflow/). | tensorflow-1.15-gpu-py37-cu110-ubuntu18.04-v8 | Python 3 | Python 3.7 |