Page View Time Series Visualizer

Page View Time Series Visualizer is a Data Analysis with Python project from freeCodeCamp. A boilerplate and instructions are found at https://www.freecodecamp.org/learn/data-analysis-with-python/data-analysis-with-python-projects/page-view-time-series-visualizer

The file fcc-forum-pageview.csv contains a time series data of the freeCodeCamp forum page views. From the CSV file, we use matplotlib and seaborn libraries to:

1. Produce a line plot of the Daily freeCodeCamp Forum Page Views 5/2016-12/2019
2. Create a bar plot of the average daily views each month, for every year available.
3. Create a box plot showing the trend of page views (views by year), and a second box plot showing the seasonality of page views (views by month)

In order to do the tasks above, the data was:

• imported from the CSV file
• dates were parsed
• dates were used as the index
• data was cleaned to remove the upper and lower 2.5 quantiles

Additional formatting were performed to:

• ensure that months were in order (instead of alphabetical)

Project Ouput:

1. Line plot of the daily page views from May 2016 to Dec 2019:

2. Bar plot of the average daily views for each month between May 2016 and Dec 2019:

3. (Left) Year-wise box plot shows increasing page views over the years and (Right) month-wise box plot shows the seasonality of page views spanning a year

---

Notes: I was encountering a numpy Attribute Error. The boilerplate version that we forked was using seaborn 0.9.0. To fix the issue, I updated the "requirements.txt" to use seaborn 0.13.2 and updated seaborn in gitpod.

I was aided by the freeCodeCamp forum: https://forum.freecodecamp.org/t/boxplot-error-for-page-view-time-series-analyzer/683684

prior to forum consultation, I used "monkey patches" as suggested here: https://stackoverflow.com/questions/74844262/how-can-i-solve-error-module-numpy-has-no-attribute-float-in-python