Jupyter Notebook Hands On Instructions

1. Go to https://demo..airavata.org/ and log in through CILogon

2. Click "Launch a Notebook" button and create a notebook server by providing a name for it. This will spin up a Jetstream2 cloud-hosted notebook server

3. Once the notebook server is available, you should be able to see a start button in front of the notebook entry

4. Click the start button and it will open up a new tab with access to the newly created notebook server

5. Click and open Example.ipynb notebook.

6. Run following code blocks in individual notebook cells

Code Block 1

with open('output-onlinerandomtools.txt', encoding='utf-8-sig') as f:
    content = f.read()
array = content.split()
numbers = []
for item in array:
    numbers.append(int(item))

Code Block 2

total_sum = sum(numbers)
print(total_sum)

Code Block 3

import matplotlib.pyplot as plt
plt.hist(numbers)

7. Above import will throw an error because the notebook runtime does not have matplotlib installed by default. To install matplotlib, get a new cell and run the following command

pip install matplotlib

8. Once the library is installed, try running code block 3 to make sure that the library is properly installed

9. Save the notebook and run the last line of a notebook with an appropriate description

Code Block 4

%export_states createArchive=True, uploadServer=http://172.17.0.1:8080, archiveName=Experiment with total sum <SUM>

10. Download the ARCHIVE.zip file to your local disk by clicking the link provided

11. Run a docker container with the downloaded ARICHIVE.zip as an input. Update the <Local Archive Path> value with the absolute path of the downloaded ARCHIVE.zip file

docker run --cap-add=SYS_PTRACE -it -p 8888:8888 -v <Local Archive Path>:/opt/ARCHIVE.zip -e NOTEBOOK_TOKEN=123456 dimuthuupe/ipykernel:1.2