TypeError: unsupported operand type(s) for -: 'str' and 'float' python pandas

Question

I have a dataframe I usesd groupby to get the table below

    time_groups=time_data.groupby('time_binned').mean()
    time_groups

Then I try to build a bar chart to show the 'Status' column result, but I got error TypeError: unsupported operand type(s) for -: 'str' and 'float'

plt.bar(time_groups.index.astype(str),100*time_groups['Status'])

I think you shouldn't pass strings as the first argument to plt.bar. Instead use something like range(len(time_groups.index)) and adjust the labels later on via plt.xticks. — a_guest
– a_guest, Commented Aug 20, 2018 at 21:47

Vishnu Kunchur · Accepted Answer · 2018-08-20 22:05:43Z

3

The reason you see the error is because the first argument (x) is required to be numeric. This is because you're specifying the x-coordinates of the plot. Try doing this:

plt.bar(x = np.arange(0, len(time_groups['Status'])), height = time_groups['Status'])
plt.xticks(np.arange(0, len(time_groups['Status'])), time_groups.index.values)
plt.show()

answered Aug 20, 2018 at 22:05

Vishnu Kunchur

1,73611 silver badges10 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

qing zhangqing Over a year ago

Hi Vishnu, it shows another error bar() missing 1 required positional argument: 'left'

qing zhangqing Over a year ago

Quick question Kunchur, what about if I want to use time_binned column data as my xsticks, do you know what will be the good method to do that?

qing zhangqing Over a year ago

It solved, I used the code you worte above and it works. Thank you very much.

L Tyrone · Accepted Answer · 2024-06-14 09:15:32Z

0

I have found the solution. Instead of the single groupby statement, this is how you do it:

Group by 'State' and sum only the numeric columns:

df2 = df.groupby('State')[['Deaths', 'Cases']].sum().reset_index()

If you still want 'County' information (for example, count of unique counties per state), you can add it separately:

df_counties = df.groupby('State')['County'].nunique().reset_index()
df_counties.rename(columns={'County': 'Unique_Counties'}, inplace=True)

Merge the two dataframes to get a single dataframe with all the desired information:

df2 = df2.merge(df_counties, on='State')

edited Jun 14, 2024 at 9:15

L Tyrone

8,36123 gold badges34 silver badges47 bronze badges

answered Jun 14, 2024 at 6:41

Iwy

12 bronze badges

2 Comments

OCa Over a year ago

What did you use for input data prior to groupby? OP hasn't even shared it ; this was a rather poorly-formulated question to begin with

OCa Over a year ago

Also how does your method affect the required subsequent bar plot, which you do not mention?

Collectives™ on Stack Overflow

TypeError: unsupported operand type(s) for -: 'str' and 'float' python pandas

2 Answers 2

3 Comments

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related