如何在平均值和置信区间误差条 python 之上添加计数文本？

Question

I have created an errorbar plot in matplotlib with the mean and confidence intervals of each bin. Now, I am trying to add a text label above each errorbar that states the number of observations in each bin. So far I have:

   binned   bin_count  val_mean  val_ci 
(0.1, 0.3]    10        3.13    14.20   
(0.3, 0.6]    40        -.1     12.98   
(0.6, 0.9]    31        1.8     12.59   
(0.9, 1.2]    4         .42.    1.42

bar1 = plt.errorbar(x = df.binned, y = df.val_mean,
             yerr = df.val_ci, marker = 'o', linestyle = '', capsize = 4)

i = 0
for line in bar1.lines:
    plt.text(x = line.get_xdata(), y = line.get_ydata()+ .1, s = str(df['bin_count'][i]))
    i += 1
plt.show()

But I keep getting the errors:

AttributeError: 'tuple' object has no attribute 'get_xdata'

and

TypeError: only size-1 arrays can be converted to Python scalars

I have been looking at documentation on how to access the x and y values of Line2D object and what I have says get_xdata().

Also when testing out what happens when printing get_xdata() it outputs the bins like so:

for line in bar1.lines: 
     print(line.get_xdata())

Output:

 ['(.1, .3]', '(0.3, 0.6]', '(0.6, 0.9]', '(0.9, 1.2]']
 AttributeError: 'tuple' object has no attribute 'get_xdata'

I'm not sure what to do to correctly access the coordinates of each bar and print the bin_count text on top of the error bars. Any help greatly appreciated.

Update: Adding dataframe (df) code

df = pd.DataFrame({'binned': ['(0.1, 0.3]', '(0.3, 0.6]', '(0.6, 0.9]', '(0.9, 1.2]'],
                   'bin_count': [10, 40, 31, 4],
                   'val_mean': [3.13, -.1, 1.8, .42],
                   'val_ci': [14.20, 12.98, 12.59, 1.42]})  

Answer 1

Is this something you are trying to achieve?

import matplotlib.pyplot as plt
import numpy

df = numpy.array([[
    0.1, 0.3, 10, 3.13, 14.2], [
    0.3, 0.6, 40, -0.1, 12.98], [
    0.6, 0.9, 31, 1.8, 12.59], [
    0.9, 1.2, 4, 0.42, 1.42]])

bar1 = plt.errorbar(df[:, 0], y=df[:, 3], yerr=df[:, 4], marker='o', linestyle='', capsize=4)


xx = bar1.lines[0].get_xdata()
yy = bar1.lines[0].get_ydata()
for i in range(len(xx)):
    plt.text(x=xx[i], y=yy[i] + .1, s=str(df[:, 2][i]))
    i += 1
plt.show()

You can make the necessary changes on your df and it should still work in my opinion. The end result I achieved with this is shown here.

Now note that, in your case I assume your first column as an interval, but you can notice in your print statement that your interval is a string. I haven't invested any time in solving your interval problem as this question is intended for having bin counts as text.