center-for-health-data-science / pythontsunami Goto Github PK

Request was to add x-axis labels in plain matplotlib in the section "Combining Matplolib, Pandas and Seaborn".

There is already an example for the seaborn plot using matplotlib.axes.Axes.set_xticklabels

To add:

_ = ax.set_xticks(ticks=x)
_ = ax.set_xticklabels(labels=list(some_counts["Denmark"].keys()), rotation=45)
# _ = plt.xticks(ticks=x, labels=list(some_counts["Denmark"].keys()), rotation=45)

the link to colab data structures in the branch fall2021 is broken

Should we document object methods or module functions using obj.method (module.function) or obj.method() (module.function()).

I prefer the former, Rita used the ladder.

I have seen both in the official documentation. I just think it will be good to be consistent.

Any opinion or good reference to elaborate the issue?

I believe it could be intuitive for health (data) researchers if lists and arrays and so on contains some kind of biological information, so that it is easier to understand/imagine what it could be used for.

For example in the Numpy exercise:

now: values = [4, 9, 1, 0, 8, 3, 2, 2, 6, 5, 0]
suggestion: amino_acids = ['A','C','H','L','R',...]

So exercise "Exercise 2: Access items", could be about removing essential amino acids or something like that?

I tried to find the origins of the vaccines dataset, I guess it is from here

• Would be maybe good to have an external ressource with updates and description for each, I guess...

Collection of questions on list

• inplace manipulation using most list methods. How to copy a list?

Code-Snippets

## List copies

a = 40
another_list = [6, 4, 1, 2, 5, a]
list2 = list(another_list) #.copy()
another_list.sort()

print(another_list)
print(list2)




## Nested list copies

# create a list
inner_list = [40, 50, 60]


outer_list = [6, 4, 1, 2, 5, inner_list]
copy_outer_list = list(outer_list) #.copy()
# copy_outer_list = outer_list.copy()

print(outer_list)
print(copy_outer_list)


inner_list[2] = 500
outer_list[1] = 100
print(outer_list)
print(copy_outer_list)

The part about formatting (variables_data-types) could be made simpler by only including one or at least not the depreciated way.

"The old way => % operator (deprecated)"

ipynb files link to https://github.com/Center-for-Health-Data-Science/PythonTsunami/blob/intro/figures/tsunami_logo.png?raw=1

but should really link to https://github.com/Center-for-Health-Data-Science/PythonTsunami/blob/intro/figures/tsunami_logo.PNG?raw=1

The python introduction (notebooks prefixed with 1_) does not introduce dictionaries (dicts), tuples or sets. It introduces the list type extensively. As both pandas and scikit-learn at least partly use or return dicts (and one exercise of numpy gets easier with dicts), I would suggest to add a brief, final notebook in the intro part introducting dicts.

The list.ipynb could be complemented with a distinction between lists and tuples.

All notebooks are also presentations using rise to render them. Add short documentation.

Checking some details, I visualized the latest plotly notebook. The plotly plots are rendered... I wasn't aware of that.

Could be added as a hint to the README.md

check yourself