olgashldrr / py_ladies_bootcamp_capstone Goto Github PK

Hi Team 5,

Congratulations on completing the bootcamp. Awesome! 👍 I found you all did a great job overall on the Notebook. Interesting dataset also 😄

Here is my feedback on the Capstone:

• Question 1 Would be useful to sort like it is done in Question 2 using sort_values(). The data could be visualized using diagrams, as is done for the other questions in the Notebook.
• Question 3 Great coding! 👏
• Question 5 It is possibly clearer for the reader if the plotting line slices the columns using names (would also be interesting to consider which of the two ways of accessing columns has better performance)
  plt.plot(tunisia['Year'], tunisia['Overweight'])
• Question 6 Clear visualization to have included a heatmap plot. In the lmplot the x-axis labels using text would be more human-readable (see more feedback on this plot below).

Here feedback about the choices made:

• Question 6 It would also have been useful to show the absolute values of the correlations:
  corrMatrix = df.corr().abs()
  This to more easily compare the impact of income on malnutrition.
  A regression line is fit in the lmplot, but what is its meaning? Instead, a box plot seaborn.boxplot or a scatter plot where one variable is categorical seaborn.stripplot could be used. If using the stripplot, the means (or medians) could be shown on top of it. These plots show more clearly for example the outliers and the number of data points.

I also have feedback on the presentation, which I hope is useful:

• Mention sources of information. For example, for the definitions of malnutrition.
• The mentioned types of malnutrition do not exactly correspond to those in the dataset. The type micronutrient deficiency is mentioned, but it is not included in the dataset. The dataset includes Wasting as well as Severe Wasting.

It was interesting to read the Notebook and presentation! 😄 Do let me know if you want to discuss further. Great job! 👍