Getting Started

The following example shows how to apply BFASTMonitor on a medium-sized dataset.

import os
import wget
import numpy
from datetime import datetime

# download and parse input data
ifile_meta = "data/peru_small/dates.txt"
ifile_data = "data/peru_small/data.npy"

if not os.path.isdir("data/peru_small"):

if not os.path.exists(ifile_meta):
    url = '', ifile_meta)
if not os.path.exists(ifile_data):
    url = '', ifile_data)

data_orig = numpy.load(ifile_data)
with open(ifile_meta) as f:
    dates ='\n')
    dates = [datetime.strptime(d, '%Y-%m-%d') for d in dates if len(d) > 0]

First, a dataset is downloaded and stored in the ‘data’ directory (if the files do not exist yet). Afterwards, a Numpy array containing the satellite time series data is loaded as well as a text file that contains the dates for the satellite images (i.e., a datetime index for the first dimension of ‘data_orig’)

from bfast.monitor.utils import crop_data_dates
start_hist = datetime(2002, 1, 1)
start_monitor = datetime(2010, 1, 1)
end_monitor = datetime(2018, 1, 1)
data, dates = crop_data_dates(data_orig, dates, start_hist, end_monitor)
print("First date: {}".format(dates[0]))
print("Last date: {}".format(dates[-1]))
print("Shape of data array: {}".format(data.shape))

Next, the start of the history period as well as the start and end of the monitoring period are defined. Given these datetimes, the data array and the dates are “cropped” .

from bfast import BFASTMonitor

model = BFASTMonitor(
        ), dates, n_chunks=5, nan_value=-32768)

print("Detected breaks")
# -2 corresponds to not enough data for a pixel
# -1 corresponds to "no breaks detected"
# idx with isx>=0 corresponds to the position of the first break

Finally, the BFASTMonitor model is defined and fitted using the ‘opencl’ backend. The data array is processed in 5 chunks.