CODEWITHSUNDEEP
pythonmatplotlibpandas
prooffreader
prooffreader

Reputation: 2443

How to pass a list of pandas columns to a pyplot chart?

Here is my code, which works fine as far as it goes:

df = pd.DataFrame({
    'foo' : [1, 2, 7, 2],
    'bar' : [3, 1, 3, 2],
    'spam' : [5, 2, 1, 0]
    })

x = range(len(df.foo))

fig, ax = plt.subplots()
ax.stackplot(x, df.foo, df.bar, df.spam)
# plt.savefig('stackedarea.png')
plt.show()

Stacked area chart

My question is, how can I pass a list so I don't have to explicitly type out every column (Df.foo, df.bar...)?

I'm still a rank beginner at lambda functions and list comprehensions, I suspect one or the other is required.

(1) My first idea, to pass a list of column names

columnlist = ['foo', 'bar']
# snip
ax.stackplot(x, #something_goes_here) #I tried df[columnlist[, no joy

(2) My second idea, to pass a list of columns:

columnlist = ['foo', 'bar']
#here is where I don't know how to transform column list so it becomes
# passedlist, where passedlist = [df.foo, df.bar]
# snip
ax.stackplot(x, df[columnlist])

Hopefully I've explained well enough. I've only been doing python for a few weeks, please don't snicker out loud!

Upvotes: 1

Views: 1595

Answers (1)

behzad.nouri
behzad.nouri

Reputation: 77951

if you want to plot all the columns you can do:

ax.stackplot(x, *[ts for col, ts in df.iteritems()])

if only a subset:

ax.stackplot(x, *[df[col] for col in ['foo', 'bar']])

note the * in above lines.

edit: you can also pass a 2-dimensional array to stackplot, so the simpler notation is:

ax.stackplot(x, df.T)  # all columns
ax.stackplot(x, df[['foo','bar']].T)  # only foo & bar columns

Upvotes: 6

Related Questions