EDA code quality

[hsmohammed]

• Try to avoid repitions in the code and make use of functions that avoid that. An example of that this chunk that you used:

  sns.histplot(train_df1, x="Length (mm)", kde=True, bins=15, ax=axs[0, 0])
  sns.histplot(train_df1, x="Diameter (mm)", kde=True, bins=15, ax=axs[0, 1])
  sns.histplot(train_df1, x="Height (mm)", kde=True, bins=20, ax=axs[0, 2])
  sns.histplot(train_df1, x="Whole Weight (g)", kde=True, bins=15, ax=axs[0, 3])
  sns.histplot(train_df1, x="Shucked Weight (g)", kde=True, bins=15, ax=axs[1, 0])
  sns.histplot(train_df1, x="Viscera Weight (g)", kde=True, bins=15, ax=axs[1, 1])
  sns.histplot(train_df1, x="Shell Weight (g)", kde=True, bins=15, ax=axs[1, 2])

  Check the link below for a better alternative to doing that. https://seaborn.pydata.org/examples/faceted_histogram.html

[chuangw6]

Hi @hsmohammed, Thank you so much! great feedback. I didn’t pay attention to the duplicated code last week. I have made changes here and the repo is updated.

Thank you for sharing the resource. As we intend to draw distribution for each column, the faceted histogram might not work here. Therefore, I use the for loop to draw the histogram for each column. This reduces the duplicated code and follows the DRY principle. I hope this is okay.

Thank you again for your time.