Automatic, range-aware formatting of time-axis labels for timeseries data using matplotlib.
This package is supported for Python 3 only. This project works for time ranges on the order of milliseconds or millenia, and produces a decent-looking result with a few lines:
from autoformat import autoformat
autoformat.scale(start_date, end_date)
# your matplotlib plotting code here
pip install autoformat
import math
import datetime
import matplotlib.pyplot as plt
from autoformat import autoformat
def get_test_data():
times = []
data = []
start_time = datetime.datetime.utcnow()
for x in range(0,20):
next_time = start_time + datetime.timedelta(hours=x)
times.append(next_time)
data.append(math.cos(x / 10.0))
return times, data
def plot(filename):
times, data = get_test_data()
start = times[0]
end = times[len(times)-1]
autoformat.scale(start, end)
plt.plot(times, data)
plt.savefig(filename)
plt.close()
plot("sample.png")