poor melt example in documentation

[MLopez-Ibanez]

Problem description

The example chosen to illustrate melt

https://pandas.pydata.org/pandas-docs/stable/user_guide/reshaping.html#reshaping-by-melt

is not ideal. Normally, melt would be used to convert from wide to long data. However, the data on the left hand side of the example is already in long format (each variable corresponds to a different column) and the melt command is just creating a strange "thing" where the column value contains two different variables. I'm using this figure in my teaching as an example of what NOT to do when reshaping data.

A better example would be (taken from: https://www.jstatsoft.org/article/view/v059i10)

first	last	Treatment A	Treatment B
Jane	Doe	16	11
Mary	Johnson	3	1

melted into:

first	last	variable	value
Jane	Doe	Treatment A	16
Jane	Doe	Treatment B	11
Mary	Johnson	Treatment A	3
Mary	Johnson	Treatment B	1

I'm happy to provide a pull request against the docs if this is considered an improvement.

[gfyoung]

cc @jreback

[MLopez-Ibanez]

Are the sources of the figures already available somewhere? If not, would it be worth to store the SVG sources in git so that they can be easily edited in the future?

[gfyoung]

@MLopez-Ibanez : Have a look at this directory:

https://github.com/pandas-dev/pandas/tree/11a26becdf98daef36178683f24b3ee105187e67/doc/source/_static

[MLopez-Ibanez]

Those do not seem to be source files (I would expect either a script to recreate the files or svg or some other source format)

[gfyoung]

Ah, sorry, I don't believe we have said files unfortunately...

[MLopez-Ibanez]

Somebody created the images. Perhaps they have the source files? It would be useful to know at least the fonts and font sizes used, to be consistent.

[MLopez-Ibanez]

• Another issue is that the figure says df3 but the example uses cheese.

From the closed PR:

Something like a dataframe for GDP, having the id column as a country, and then having columns 2016, 2017 and 2018 with the values could be a good example, easy to understand. Made up values in A, B columns is not.

(Nevertheless that most examples in that page use "A" and "B" as columns...)

The original assets are available here: #28006 (comment)

but they are in non-free formats.