Skip to Content

Set Up Jupyter Notebook for Service Ticket Intelligence

test
0 %
Set Up Jupyter Notebook for Service Ticket Intelligence
Details

Set Up Jupyter Notebook for Service Ticket Intelligence

July 27, 2020
Created by
July 15, 2020
Set up a local JupyterLab instance as well as the Jupyter Notebook for Service Ticket Intelligence.

You will learn

  • How to install Python and Anaconda
  • How to set up a local Jupyter server to run Jupyter notebooks

A Jupyter Notebook is used in the following tutorials to make the steps interactive and easy to follow. For more information on Jupyter Notebooks, see Jupyter. The notebooks and datasets used for this tutorial mission are available in this repository.


Step 1: Install Anaconda

First, you need to install Anaconda. Anaconda is a platform that offers tools to process large datasets and is often used by data scientists. When installing Anaconda, the programming language Python is installed as well.

To install Anaconda, head over to the Anaconda installation guide and look for your operating system. Once there, follow the installation guide.

To verify the installation, search for Anaconda Prompt in your programs and open it. In the command prompt, enter conda --version and the version is printed as in the image below.

To verify the installation, open a new Terminal or Command Prompt. In the command prompt, enter conda --version and the version is printed as in the image below.

Check Anaconda Version

You have successfully installed Anaconda.

Log on to answer question
Step 2: Clone the repository

All you will need throughout the tutorial is within this repository on GitHub. It includes two Jupyter notebooks, one for each of the use cases of Service Ticket Intelligence, as well as some dataset that can be used. In the following, you will clones this repository to your local computer.

Open the Anaconda Prompt again in case you have closed it. Navigate into the folder where the repository should be placed using the command cd <path> or stay where you are. Then, enter the following command to clone the repository:

git clone https://github.com/SAP-samples/service-ticket-intelligence-jupyter-notebook-sample

Open a Terminal or Command Prompt again in case you have closed it. Navigate into the folder where the repository should be placed using the command cd <path> or stay where you are. Then, enter the following command to clone the repository:

git clone https://github.com/SAP-samples/service-ticket-intelligence-jupyter-notebook-sample
Clone Repository

Once the repository is cloned, you receive an output as above, saying that it’s done.

Log on to answer question
Step 3: Configure service credentials

Next, you need to configure the notebooks to use the service key that you created in Create Service Instance for Service Ticket Intelligence.

Open a file explorer and navigate to the folder called service-ticket-intelligence-jupyter-notebook-sample that was just created. Open the file called sti_config.ini in a simple text editor.

Configuration File

Then, insert the following values from your service key into the corresponding spaces of the template. Make sure to copy only the values, without the surrounding quotes.

Service key property Property in template
sti_service_url (outside the uaa section of the service key) service_url
url (inside the uaa section of the service key) uaa_url
clientid client_id
clientsecret client_secret

Finally, save the file.

Log on to answer question
Step 4: Start Jupyter server

To use the Jupyter notebooks, you must have a local instance of the Jupyter server running.

Stay in the Anaconda Prompt (Windows) or the Terminal (Mac and Linux) and navigate into the new folder that was created while cloning the repository by using the command cd service-ticket-intelligence-jupyter-notebook-sample. You then move into the folder and your current path will change accordingly.

Changing Folders

Next, enter the command jupyter notebook to start the server.

Start Server

Once the server started, the site automatically opens in browser. In case it does not, you can use the URLs that were printed in the command prompt.

With that, you are done with setup and can proceed with the following tutorials.

Log on to answer question
Step 5: Test yourself
Select the correct command to start your Jupyter server:
×

Next Steps

Back to top