Notebook environments (Watson Studio)
With Jupyter notebook environment definitions for Python and R, you can select the hardware and software configuration of the runtime environments for notebooks that you run in the notebook editor.
- Default Jupyter notebook environment definitions
- Other environment options for notebooks
- File system in Jupyter notebook environments
- Runtime scope
Default Jupyter notebook environment definitions
You can use the default Jupyter notebook environment definitions for Python and R to quickly get started with notebooks, without having to create your own environment definitions. The default environment definitions are listed on the project’s Environments page.
The default Python environments are included with Watson Studio.
The Default Python 3.7 environment includes the latest open source versions of many popular machine learning libraries like TensorFlow, XGBoost and PyTorch. The old Default Python 3.7 environment was renamed as Default Python 3.7 (legacy).
| Name | Hardware configuration |
|---|---|
Default Python 3.7 |
1 vCPU and 2 GB RAM |
Default Python 3.7 (legacy) * |
1 vCPU and 2 GB RAM |
Default Python 3.6 * |
1 vCPU and 2 GB RAM |
* This environment is deprecated. Start running Python notebooks in environments with Python 3.7.
If you also have the Jupyter Notebooks with R 3.6 service installed, the default R environment is listed.
| Name | Hardware configuration |
|---|---|
Default R 3.6 |
1 vCPU and 2 GB RAM |
Other environment options for notebooks
You can create custom notebook environment definitions. See Creating environment definitions.
If you are coding Python notebooks or scripts in the JupyterLab IDE, you can use a JupyterLab environment. See JupyterLab environment definitions.
If you are creating scikit-learn, XGBoost, PyTorch, TensorFlow, Keras, or Caffe models, or are coding Python functions or scripts in Watson Studio notebooks, select to use the software version Default Python 3.7 in the environment definition.
If you have the Execution Engine for Apache Hadoop installed, you can create Hadoop environment definitions to run notebooks on your Hadoop cluster. See Hadoop environments.
If you have the Analytics Engine Powered by Apache Spark service installed, you can choose from default Spark environment definitions with multiple hardware configurations for Python, R, and Scala. See Spark environments.
If you have the Jupyter Notebooks with Python 3.7 with GPU service installed, you can create an environment definition to run notebooks on GPU clusters. See GPU environments.
File system in Jupyter notebook environments
You must be mindful of the size of the data files you load to your notebook. Very large files might require more storage than is available on the node on which the runtime is started.
Be aware that the file system of each runtime is non-persistent and cannot be shared across environments.
Runtime scope
Environment runtimes are always scoped to an environment definition and a user within a project.
For example, if you associate each of your notebooks with its own environment, each notebook will get its own runtime. However, if you open a notebook in an environment, which you also selected for another notebook and that notebook has an active runtime, both notebook kernels will be active in the same runtime. In this case, both notebooks will use the compute and data resources available in the runtime that they share.
If you want to avoid sharing runtimes but want to use the same environment definition for multiple notebooks in a project, you should create multiple custom environment definitions with the same specifications and associate each notebook with its own definition.