13 -- customer complaints -- Will Butts

[Shea-Wolfe]

• [x] Blank slate
  • [x] Create a GitHub repo called consumer-complaints
  • [x] Put README.md and requirements.txt files in it
  • [x] Make sure any necessary libraries are in requirements.txt
  • [x] Make sure your consumer-complaints folder has its own virtual environment (.envrc), but do not commit .envrc or .direnv
  • [x] Install the requirements using pip install -r requirements.txt
• [x] Normal mode
  • [x] Read in the data file
  • [x] Calculate and chart:
  • [x] Number of complaints by product
  • [x] Number of complaints by company (top 10)
  • [x] Number of complaints by company response
  • [x] Mean number of complaints by day of the week
  • [x] Any other insights you find interesting
  • [x] Summary text for each chart
• [ ] Hard mode
  • [ ] Combine the complaints data with the population of each state and chart the frequency of complaints by state per capita
  • [ ] Find statistically significant outliers of # of complaints by ZIP code and try to find a possible reason for the outliers
• [ ] Nightmare mode
  • [ ] Make an infographic of a map of the US, showing the frequency of complaints per capita using shading of each state, with lighter colors representing low frequency and darker colors representing high frequency

[Shea-Wolfe]

@jamesmallen @powder-river

https://github.com/Shea-Wolfe/consumer-complaints

I'm not entirely happy with my solution for the last problem but it dynamically scales so I'll take it.

[powder-river]

@Shea-Wolfe is see what you are saying with your solution, however on my end it appears you did in fact find the mean so good on you!