如何格式化Matplotlib辅助y轴日期时间值

Question

I am attempting to plot a line graph superimposed on a bar graph using matplotlib. The data has plotted correctly, however, the secondary y-axis describing the forecasted start time is being output incorrectly. All values for Forecasted Start Time are rounded off to the hour eg (13:00, 14:00, etc) but when plotted are converted to 11:40, 12:13:20, etc on the y-axis, as so:

Forecasted Start Time csv data

Forecasted Start Time and Event Length vs. Forecast Run graph

The code to produce this image is as follows:

en1 = []
en2 =[]

en1 = np.array(fh.loc(axis=0)[0:44])
en2 = np.array(ev.loc(axis=0)[0:44])

fig,ax = plt.subplots(figsize=(10,10))
ax.bar(en1,en2,color='gainsboro')
plt.ylabel('Forecasted Event Length')

plt.xticks(rotation='90')
ax2 = ax.twinx()
mn, mx = ax2.get_ylim()
ax2.set_ylabel('Forecasted Start Time')
color = 'tab:blue'
ax2.tick_params(axis='y')


ax2.plot(fh,start, marker='o')

plt.legend()

Any ideas on how to plot the actual datetime values from the csv correctly on the secondary y-axis would be greatly appreciated!

Answer 1

You need to plot on the axis object using ax.bar and ax2.plot instead of plt.bar and plt.plot .

You can use the axis formatter to do this:

from matplotlib.dates import DateFormatter

ax2.yaxis.set_major_locator(md.HourLocator(interval=1))
ax2.yaxis.set_major_formatter(md.DateFormatter('%H:%M'))