matplotlib formatting x axis with timestamps from big data

Asked Sep 20 '18 at 16:12

Active Sep 20 '18 at 16:12

Viewed 174 times

I am trying to create a plot that has a lot of data on it. This one in particular has about 550 points on it, each with its own timestamp. When I plot this, there are so many timestamps that I just get a black bar. I know it is not reasonable to expect to be able to make all timestamps visible, but is there a way to format the ticks t=so that they represent the range of values?

Here is my code:

plt.figure(1)
plt.scatter(x_axis_input, y_axis_input, s=DOT_SIZE)
plt.xlabel('timestamp')
plt.ylabel('value')
plt.title('test')
plt.savefig('plot_test.png')
plt.close()

and here is the resulting plot:

Link to plot

asked Sep 20 '18 at 16:12

griffer98

you can rotate the timestamps on x axis to get a better view – Atul Shanbhag Sep 20 '18 at 16:14
Rotating 550 stamps won't help either – Sheldore Sep 20 '18 at 16:15
if there are 550 stamps, you can display only say 10 stamps and have them rotated – Atul Shanbhag Sep 20 '18 at 16:16
https://stackoverflow.com/questions/14852821/aligning-rotated-xticklabels-with-their-respective-xticks/14854007#14854007 https://matplotlib.org/gallery/ticks_and_spines/ticklabels_rotation.html – Atul Shanbhag Sep 20 '18 at 16:17
the solution in the above link does not work – griffer98 Sep 20 '18 at 17:03
Matplotlib automatically uses nice spacings for the ticklabels if they are really datetime objects or timestamps. If you see a large black bar, it means that you are not using numeric data, but strings. Convert your strings to datetime to solve this. – ImportanceOfBeingErnest Sep 20 '18 at 22:51

matplotlib formatting x axis with timestamps from big data

0 Answers0