Good Day All! Would Love Feedback for the eBay Car Sales Project

Hi folks. I’m really looking forward to some feedback on this project. I’ve worked on style, comments and improved code, but would love to receive pointers on how to improve readability and code structure if possible. My last project (Hacker News) reviews from @Edelberth and @jesmaxavier were really helpful. I look forward to continuing this process. Thank you for your feedback.

Ebay_Car_Sales_Germany.ipynb (116.6 KB)

Click here to view the jupyter notebook file in a new tab

3 Likes

Hi @aarong98104
Thanks for sharing your project with Dataquest community. The project looks great with the title, introduction part, the comments, explanations given in the markdown cell very informing .keep it up mate for the good work. To me everything looks great, however ;

  • Always consider having one block of code line as oppose to multiple lines with same code,
  • you should also do more on explanation/observations, workings from cell[146] to cell[157] are somehow lengthy and without explanations or observation made, it will be very difficult for users to understand what is really happening more so those with little knowledge in programming.

Otherwise ,congratulations mate for the good work. Looking forward to interact more of your projects. All the best in your upcoming projects.

1 Like

Thank you @aarong98104 for having named me, the truth is that it is a very good detail on your part, I am no eminence in the field, so let’s see if I can serve you.

First thing I see is that the questions you want to answer with your analysis do not appear. nor do the meanings of the columns appear, so someone who is not from DQ will not understand many of the things we take for granted.

Another is that you use bold and it is a format that to highlight helps but reading everything so tires the eyes :eyes: because is black and you must pay attention to reading in a more intense way.

Use markdown to point out points where you want to summarize things. For example:

column_1 column_2 column_3
1 2 3
blah blah blah
etc etc

Explore the possibility of showing different columns with pandas, helps to clarify things and relate ideas in addition to not having an go step by step with each line of code.

Select a list of rows

list_rows = f500_selection.loc[["Toyota Motor", "Walmart"]]  <- more than 2 columns!!! 
print(type(list_rows))
print(list_rows)

class 'pandas.core.series.Series'

rank             3
revenues    267518
profits     1257.9
country      China
Name: Sinopec Group, dtype: object

I hope I have helped you, what helps me personally is to imagine people who do not know anything and make the effort of how I would explain a story with the things that I know so far.

if you have any questions, do not hesitate to write to me.

A&E.

1 Like

A big thank you to @Edelberth and @brayanopiyo18 for giving me pointers. On my next project I’m going to focus on attempting to make readability for someone outside of DQ. I think sometimes I’m so wrapped up in trying to answer the outline of the project that I forget what it looks like for readers. As to utilizing more consolidated and specific code, I believe I’m going to have to revisit some of the lessons again. I look forward to following more of your projects and sharing some of mine. It means a lot to me for people to offer honest and helpful feedback. Talk to you soon and good luck with your work!

2 Likes

Thank you.

it also happened to me, I loss the focus and spot on the details of the coding.

What I do is follow the instructions of DQ once I understand why they have decided this or that then with everything I have been seeing I try to add something of value and if not look at what others have done.

Here, going over again is not wasted time.

Can I send you a message next time to take a look at something I’ve done?

See you here…

A&E

@aarong98104 congrats :handshake: on completing your project. You’ve gone quite deep in to analysing the columns. Keep this exercise going, it’ll be very useful when you get into the Data Science portion of this course.

Below are a few pointers that I have:

Presentation Style
  • Once you feel you are done with the project, re-run the entire project so that the cells start from [1]. This makes it easier for reviewers like me to refer to cells. image
  • It would be good to include a couple of sections like Introduction, Reading the Data, Cleaning the Data etc. It would help the project look more formal.
  • You’ve emboldened the entire text portion of your project. I would avoid this. Instead, I recommend you embolden only those portions of the text that may be relevant to the reader. Consider the following example after cell [165]:

The top 10 brand models contain nine German automakers, with Mercedes and Volkswagen coming in with over 12.5% of popular models.

  • I would also encourage you to use bullet points instead of full paragraphs, as you have done in the paragraph after cell [102]. The following is what I did for the same part in my project:

    TL;DR readers may find this helpful
  • Just as your last project, this one is also heavy on terminal outputs. You could consider emboldening your outputs. The method for the same was provided the last time. I mention this because you tried to bring out a header for your output of cell[103].
Coding Style
  • You have included informal comments. When you go over this project in the next round, try and make them short and simple. This would help the project look clean.
  • I like what you did in cells [160] and [161] by creating brands and their respective models. My own understanding of this part was that we had to find the top model in each brand, as opposed to the top brand-model pairs.
  • In cell [103], I noticed you used a dictionary to rename columns. I would encourage you to create a function to do the same. You could use this function in later projects.
Bugs/Inaccuracies
  • There are some spelling and capitalization mistakes that I noticed that you could clear out in the next iteration.
Miscellaneous
  • In the next iteration, I would recommend that you take a second look at the price column. Your current price range starts from $100.00. I think you could increase it further if you analyse further. Hint: look for the word ‘schlachten’ in the dataset
  • Once you have got a hold on visualizations. I recommend that you re-do this project and add a couple of visualizations.

Hope that was helpful. Keep the magic :dizzy: going on!

Thanks again for the support! Suggestions are duly noted. I’ll keep an eye out for your work as well. Looking forward to continued improvement, learning and success thanks to the many helpful people on DQ.

1 Like