Exploring Ebay Car Sales Data - Winning the Binning on odometer_km column!

This guided project gave me a chance to test myself in Pandas and glad to have been able to take on everything under next steps. For the Analysis part, I didn’t follow all the instructions to the letter and changed it up a little so, the results may look a bit different from others in Analyzing mean prices across brands. I also used some pandas methods like finding the row with the MAX value of price,
constructing a dataframe from a dictionary,
And finally, Binning using pd.cut to split the odometer_km into groups to aggregate by

I wish I could add visualization to this. However, I am on that mission now.

Some feedback points I would appreciate are:

  • Should the notebook have been more concise?
  • Does my summarizing create interest?
4 Likes

Hi,

I really liked some of your functions. Especially, function converting camelCase labels to snake_case. It can also be used for Pascal convention also as you lowered first letter. Some functions like print_some_rows seems a bit superficial but I guess you are aware of that and wanted to write more code to practice. I would only add more markdown cells as it helps to follow your logic in project and conclusion. Wish you the best in your learning process :slightly_smiling_face:.

Thank you for taking the time to read through my project.

My rationale for writing a function is:

  • I want to type the least amount. This is also the reason for to_snake function as I don’t have to type column names to rename. I also learned that df.column is same as df[‘column’] and involves less typing:-)

  • With Jupyterhub helping render the DataFrame into a table which is super nice, it’s still necessary to generate the df first, using the pd.read_csv method. And same with df.head() - needs the df. Instead, I proposed myself a function that just needs my input csv and how many lines. I don’t need to type anything else.

  • I can reuse it