1

I want to use a specific module for time series break detection (called "bfast", see https://bfast.readthedocs.io/en/latest/index.html) which requires as input a 3D-numpy array.

The question is: "How I can make the Google Earth Engine ImageCollection format to fit the parameters of the bfast function?"

I am quite new to the Google Earth Engine environment so maybe a missed an obvious solution

Basically I would like to make it run on a Google Collab notebook using the Google Eath Engine API.

Here are the requirements of the bfast function I intended to use:

fit(self, data, dates, n_chunks=None, nan_value=0)
 |      Fits the models for the ndarray 'data'
 |      
 |      Parameters
 |      ----------
 |      data: ndarray of shape (N, W, H),
 |          where N is the number of time 
 |          series points per pixel and W 
 |          and H the width and the height 
 |          of the image, respectively.
 |      dates : list of datetime objects
 |          Specifies the dates of the elements
 |          in data indexed by the first axis
 |          n_chunks : int or None, default None
 |      nan_value : int, default 0
 |          Specified the NaN value used in 
 |          the array data
nmatton
  • 49
  • 7
  • 1
    You can't use Numpy functions with Earth Engine. See this post about [client vs server processing](https://developers.google.com/earth-engine/client_server). However, you can implement the bfast algorithm on EE which will probably take some fun coding. Or, you can implement the already provided [CCDC algorithm](https://doi.org/10.1016/j.rse.2014.01.011) in EE, `ee.Algorithms.TemporalSegmentation.Ccdc()` – Kel Markert Nov 20 '19 at 18:51

0 Answers0