Python Matplotlib – Bar chart on their representing sampling position on x-axis

Question

I want to plot sampling data of different groups in a stacked bar chart. The sampling was performed at different km. The distance between each sampling point is not equal. Normally, when plotting with matplotlib.pyplot.bar or pandas.DataFrame.plot.bar, the bars are plotted one behind the other. Their km-value isn't represented on the x-axis. How I can plot stacked bar plots at their representing km-position on the x-axis?

Code of a standard bar plot with pandas.DataFrame.plot.bar :

import pandas as pd
import matplotlib.pyplot as plt

df = pd.DataFrame( {'Group 1': {-60.0:0, -20.0:0, 12.5:0, 62.0:0, 123.8:0, 181.0: 5.013532366071429e-06, 225.2: 0.00010224713604266826, 248.0: 0.0002520240051269531, 274.9: 0.0006304542296807856, 304.2: 0.0009587457616051962, 331.0: 0.0021422429744175505}, 'Group 2': {-60.0: 0.0003144776457026891, -20.0: 5.43150903588747e-05, 12.5: 0.00012757662141348495, 62.0: 6.852403753623154e-05, 123.8: 5.980538377849872e-05, 181.0: 5.000001780657088e-05, 225.2: 0.00010152032391840468, 248.0: 0.0005436288535458056, 274.9: 0.00038244130009346957, 304.2: 0.00023423789360943164, 331.0: 9.508221455006986e-05}, 'Group 3': {-60.0: 0.00021804919790451726, -20.0: 0.0002884471518114942, 12.5: 0.00024001954291413006, 62.0: 0.00020780311751064946, 123.8:0, 181.0: 0.0003548555407567293, 225.2: 0.0011448858440205976, 248.0: 0.0031436022397010425, 274.9: 0.001858462242669843, 304.2: 0.0019485330483867962, 331.0: 0.0017062062250634059}} )
ax = df.plot.bar(stacked=True)
ax.set_ylabel('TM [mg/l]')
ax.set_xlabel('km')
plt.tight_layout()

For clearance:

Standard bar plot

What I want

Answer 1

With the hint of ImportanceOfBeingErnest I used matplotlib.pyplot.bar instead of pandas.DataFrame.plot.bar to get what I wanted:

import pandas as pd
import matplotlib.pyplot as plt

df = pd.DataFrame({'Group 1': {-60.0:0, -20.0:0, 12.5:0, 62.0:0, 123.8:0, 181.0: 5.013532366071429e-06, 225.2: 0.00010224713604266826, 248.0: 0.0002520240051269531, 274.9: 0.0006304542296807856, 304.2: 0.0009587457616051962, 331.0: 0.0021422429744175505}, 'Group 2': {-60.0: 0.0003144776457026891, -20.0: 5.43150903588747e-05, 12.5: 0.00012757662141348495, 62.0: 6.852403753623154e-05, 123.8: 5.980538377849872e-05, 181.0: 5.000001780657088e-05, 225.2: 0.00010152032391840468, 248.0: 0.0005436288535458056, 274.9: 0.00038244130009346957, 304.2: 0.00023423789360943164, 331.0: 9.508221455006986e-05}, 'Group 3': {-60.0: 0.00021804919790451726, -20.0: 0.0002884471518114942, 12.5: 0.00024001954291413006, 62.0: 0.00020780311751064946, 123.8:0, 181.0: 0.0003548555407567293, 225.2: 0.0011448858440205976, 248.0: 0.0031436022397010425, 274.9: 0.001858462242669843, 304.2: 0.0019485330483867962, 331.0: 0.0017062062250634059}})

width = 15
bottom = 0

for i in df.columns:
    plt.bar(df.index, df[i], width=width, bottom=bottom)
    bottom += df[i]

plt.ylabel('TM [mg/l]')
plt.xlabel('km')
plt.legend(df.columns)
plt.tight_layout()

Answer 2

You can use pyplot.plot(data = df) function from matplotlib(with default values). It will auto format xticks. Or else you can try pyplot.xticks and pyplot.set_xticks to customize your needs.