search

NumEconCopenhagen / projects-2023-s

projects-2023-s created by GitHub Classroom
MIT License
0 stars 0 forks source link

Feedback to data project #2

Open fls808 opened 1 year ago

fls808 commented 1 year ago

Very good!

  1. The most elegant solution in the project was:

    • It is very cool that the 'plot_func' is a function. Overall I like the production of that graph very much, it's elegant.
    • I like that you use 'Country' as an index.
    • It is elegant that there is 3 dimensions in the happiness, GDP, life expectancy-graph!
    • It's nice that you are so clear about causality and correlation

  2. The hardest section of code in the project to understand was:

    • I like that you scale the life_expectancy. First I was however in doubt on what the life_expectancy_sizes does. Maybe write it out explicitly, and generally provide more comments on what you do.
    • Generally it was a little difficult to understand what you do, because it is advanced. That is okay and nice, but again the assignment could benefit from more documentation.

  3. This part of the project could be better documented: (explain what)

    • Maybe remove import of packages you do not use, for example venndiagram.

  4. An idea for an improvement/clarification could be: (explain what and why)

    • I would not change the name of columns that already have the right name, for example "Region".
    • Also I wouldn't rename the columns you remove later, but I guess it's okay.

  5. An idea for an extension could be: (explain what and why)

    • I think that your project is very, very good as it is. If you should improve it you could maybe do some of the regressions you talk about and include stuff the population size. Maybe this changes the correlation, which illustrates your point about potential other drivers.
lj-username commented 1 year ago

Feedback from best team:

  1. The most elegant solution in the project was: Standardizing column names for all the datasets, using a dictionary. You transferred the different column names to a standard set, facilitating the merging of different data.

  2. The hardest section of code in the project to understand was: The dropdown widget could have had more explanation, to explain its purpose and use. For example, why use a widget instead of selecting the relevant rows/columns.

  3. This part of the project could be better documented: (explain what) For the same reason as above, the widget aspect could have been better explained. The heat map could have been better explained, and may not be the most informative or the best way to demonstrate the correlation. However, overall, I think the comments were quite informative.

  4. An idea for an improvement/clarification could be: (explain what and why) Loops could have been employed for the first section of the code. For example, when adding a new column to each dataset to store the respective year, a for loop could have been used instead of writing very similar code 5 times.

  5. An idea for an extension could be: (explain what and why) Other plots visualising the different correlations. The scatter plot was nice and informative, but I think more could be explored than just with the heat map. Perhaps exploring other correlations with data from the other columns too.