I collaborated with Deepnote to prepare the ‘Notebooks Year In Review’ report for 2020.

For the report, I analyzed 10,000 repositories on Github created in the year 2020 containing Jupyter Notebooks.

I also analyzed the Google and YouTube search trends to find the most significant queries related to Juyter Notebooks. The result I found is really interesting.

Based on the analysis, I found of these patterns in the notebooks:

  • The most used Python version is 3.6. The reason behind it could be the stability that Python 3.6 offers.

  • The top 3 imported libraries are: NumPy, Pandas and Matplotlib.

  • Matplotlib is the most famous plotting library with a clear lead over plotly and seaborn.

  • The most starred repository of 2020 is Fast AI’s Fastbook with 11k stars, 39 contributors, and 3.4k forks.

What does Google Search reveal?

I used Google Search trends to analyze the queries and topics related to Jupyter Notebooks.

TSone of the top queries on google and youtube search are:

  • jupyter notebook
  • python
  • python jupyter
  • install jupyter
  • how to use jupyter notebooks

I enjoyed analyzing Gigabytes of data science notebooks and finding the interesting patterns. Creating the dataset was a challenge but seeing the results, its all worth it.

You can read the full report here.