How do I exclude a few columns from a DataFrame plot?

Question

I have a DataFrame with about 25 columns, several of which hold data unsuitable for plotting. DataFrame.hist() throws errors on those. How can I specify that those columns sho

Answer 1

I imagine you could just:

df.drop(['bad col1', 'bad col2', 'bad col3', ...], axis=1).hist()

Answer 2

Note, a modification to @Chang She's response, as of pandas 0.16, the - operator is scheduled for deprecation. The difference() method is encouraged in its place.

exclude = ['bad col1', 'bad col2']
df.loc[:, df.columns.difference(exclude)].hist() 

---

Update on deprecation:

df - df['A']

is now deprecated and will be removed in a future release. The preferred way to replicate this behavior is

df.sub(df['A'], axis=0)

Answer 3

Following the official docs you can use loc

df.loc[:,['A','B']]

And you obtain, for instance, only the column 'A' and 'B'. In this way you can select the columns to plot.

Answer 4

We can potentially make this easier (I created a github issue), but for now you can select out the columns you want to plot:

df.ix[:, df.columns - to_exclude].hist()

Answer 5

I usually do the following:

columns = [column for column in df.columns if df[column].dtype == 'float64']
df = df[columns]

Answer 6

How about just

df_new = df[df.columns -[cols_to_exclude]]
df_new.plot()