Add time markers and see the distribution of frequency versus time. Resolution and bounds are dynamic.

The whole project is just a single, tiny class Histograph which allows you to create histograms of your data. Your data consists of timestamps. The resulting histogram will show you how many markers there were at given time.

from histograph import Histograph

# Create the class
h = Histograph()

# Lets put ten markers to the history
for i in range(10):
  marker = h.make_marker(second=i 1)
  h.add_marker(marker)

# Lets add a high spike at second 7
for x in range(5):
  m = h.make_marker(second=7)
  h.add_marker(m)

# Let's generate the histogram for our data
print(h.get_histogram(resolution=10))

The result will be:

# Notice the spike at seventh second
[1, 1, 1, 1, 1, 1, 6, 1, 1, 1] 

marker = make_marker(year=1970, month=1, day=1, hour=0, minute=0, second=0)

Converts a date and/or time into a marker. The returned marker is a unix timestamp. This function has the granularity of one second. If you need a greater precision in your histogram, then implement your own make_marker function.

marker = 3.1415
hist = Histogram()
hist.add_marker(marker)

Adds a marker to the histogram. The histograph accepts arbitraty numbers as the value of the marker. When using the make_marker function, the histograph class has a granularity of one second. If a different precision is needed, use a custom function to make the markers.

values = h.get_histogram(resolution=10, first=None, last=None)

Returns the histogram data as a list. The number of items in the list is defined by the resolution parameter. The bounding markers are defined by the first and last parameters.

The histogram data tells you how many marker indices there were in the given time between the first and last marker indices. If first is not given, the histogram will start from the "oldest" one. If last is not given, it will stop at the last marker.

first, last = h.get_timespan()

Returns the oldest and youngest markers. This allows you to mark in time the beginning and end of the histogram.