Skip to main content
After starting a notebook instance, Wherobots Cloud opens a pre-configured JupyterLab environment where you can write and execute spatial queries, visualize geospatial data, and build analytics workflows. For information on starting a runtime and creating a notebook instance, see Notebook Instance Management.
The introductory notebook appears when you first open a Wherobots Cloud Notebook. After that, you will see the JupyterLab Launcher.
JupyterLab Launcher

​
Before you start

The following is required to manage a Wherobots Notebook:
  • An account within a Community, Professional, or Enterprise Edition Organization. For more information, see Create an Account.

​
Available kernels

Wherobots Cloud provides two kernels for your Jupyter Notebooks:
  • Python kernel (ipykernel) — for Python-based spatial workflows using Apache Sedona and WherobotsDB.
  • Scala kernel (Scala) — for Scala-based spatial workflows.
To create a new notebook with either kernel, click File > New Launcher in JupyterLab and select the kernel.

​
Execute notebook cells

​
Run all cells

To execute all code cells in a notebook, do the following:
  1. Click Run in the JupyterLab toolbar.
  2. Click Run All Cells.
Execute all cells
When you first execute a WherobotsDB code cell, you may see the following warning:
WARN TaskSchedulerImpl: Initial job has not accepted any resources;
check your cluster UI to ensure that workers are registered and have sufficient resources
This is expected behavior. Executors take 1-5 minutes to start depending on the runtime size.

​
Monitor jobs with the Spark Web UI

The Spark Web UI helps you monitor running jobs, analyze query performance, and identify bottlenecks. To access it, click Sedona Spark in JupyterLab and select the correct port number.
Spark UI
To retrieve the port number programmatically, run the following in a notebook cell: 
# Get the Spark Web UI port number
spark_ui_port = sedona.sparkContext.uiWebUrl.split(":")[-1]
print(spark_ui_port)
For more information, see Web UI in the Apache Spark documentation.

​
Import custom Python modules

You can package custom Python modules as a zip file and import them into your notebook. This is useful for reusing shared utility functions across notebooks and job submissions.

​
Create and upload a module

  1. On your local machine, create a directory for your module: 
    mkdir zipmoduletest
  2. Create a Python file with your module code. For example, create zipmoduletest/hellosedona.py: 
    def hello(input):
    return 'hello ' + str(input)
  3. Add an empty __init__.py file to make the directory a valid Python package: 
    touch zipmoduletest/__init__.py
  4. Verify the directory structure: 
    ls zipmoduletest
    The output should show: 
    __init__.py       hellosedona.py
  5. Zip the module files: 
    cd zipmoduletest && zip -r9 ../zipmoduletest.zip *
  6. Upload zipmoduletest.zip to Wherobots Managed Storage or your integrated Amazon S3 bucket. For more information, see Notebook and Data Storage.

​
Use the module in a notebook

After uploading, add the zip file to the Spark context and import your module: 
# Add the zip file to the Spark context
sedona.sparkContext.addPyFile('s3://<your-bucket>/path-to-file/zipmoduletest.zip')

# Import and use the custom module
from zipmoduletest.hellosedona import hello

result = hello("Sedona")
print(result)
Output: 
hello Sedona
You can also use addPyFile to import custom Python modules in job submissions, not just notebooks.

​
Access data from Amazon S3

For information on reading and writing data from an integrated Amazon S3 bucket, see Access Integrated Storage in a Notebook.
To use new storage integrations or catalogs in your notebooks, you must start a new runtime. Notebooks can only access storage integrations or catalogs that were created before the runtime started.

​
Open a notebook by path

To open a specific notebook file, do the following:
  1. Click File > Open from path.
  2. Enter the notebook path.
  3. Click Open.

​
Export notebooks

You can export notebooks as executable scripts for use with WherobotsRunOperator job submissions.

​
Export a Python notebook

To export a Python notebook as an executable script, do the following:
  1. Click File in the JupyterLab toolbar.
  2. Hover over Save and Export Notebook As…
    Save and export notebook
  3. Select Executable Script.
    Select executable script
    The .py file will download to your machine.

​
Export a Scala notebook

To export a Scala notebook as an executable script, do the following:
  1. Click File in the JupyterLab toolbar.
  2. Hover over Save and Export Notebook As…
    Save and export notebook
  3. Select Executable Script.
    Select executable script
    The .scala file will download to your machine.
The exported Scala file does not include a main class. To make it runnable, wrap your code (excluding import statements) in an App object:
object MyApp extends App {
  // Your exported code here (excluding import statements)
}
You cannot execute .scala files in a Python kernel. Use the Scala kernel to run Scala code.
You can import the Scala executable file into sedona-maven-example/src/main/scala/com/wherobots/sedona/ for job submission.

​
Build a JAR file from a Scala notebook

To package your Scala notebook as a JAR file for job submission, do the following:
  1. Click File > New Launcher in JupyterLab.
    Open launcher
  2. Open Terminal.
    Open terminal
  3. Navigate to the sedona-maven-example directory and build the project: 
    cd sedona-maven-example
mvn clean package
  4. In the JupyterLab file browser, locate the target folder.
    Target folder
  5. Right-click sedonadb-example-0.0.1.jar and select Download.
    Download jar file
You can add custom dependencies to the pom.xml file located at notebook-example/scala/sedona-maven-example.
Once you have the JAR file, refer to WherobotsRunOperator to create a job.

​
Session expiration

A warning about a lost server connection may appear after extended use. Wherobots uses cookies for authentication that expire after one hour for security. Refresh the page to be redirected to the login screen. After logging back in, you will return to your previous page.