Quick overview¶
Here are some quick examples of what you can do with
L3AlgoProduct objects. Everything is explained
in much more detail in the rest of the documentation.
SISPPEO as a Python package¶
To begin, import SISPPEO:
import sisppeo
from sisppeo.products import L3AlgoProduct
Then, you can:
generate a L3 product (algo/mask) or a time series of L3 products
load an existing L3 product
plot stuff.
Generate a L3 product (L3AlgoProduct)¶
config = {
'input_product': filepath, # the filepath of your image
'product_type': 'S2_GRS', # which type of product it is ({sat}_{atm_corr})
'algo': 'spm-get' # which algorithm to use; this one derives SPM
}
l3product = sisppeo.generate('l3 algo', config)
See this.
Read & write SISPPEO products (netCDF files)¶
SISPPEO objects (L3AlgoProduct, L3MaskProduct, TimeSeries) are stored as netCDF files (widely used in geosciences, very handy, and the xarray.Dataset data model was inspired by it).
You can directly read and write SISPPEO objects from/to disk using the two
from_file() and
save() methods:
l3product = L3AlgoProduct.from_file('example.nc')
l3product.save('example_bis.nc')
It is common for TimeSeries not to be saved as such but to be distributed
across multiple L3AlgoProduct files (=> one per date). You can compile them
by using from_files().
from sisppeo.products import TimeSeries
timeseries = TimeSeries.from_files(['date1.nc', 'date2.nc'])
The Command Line Interface (or CLI)¶
(Almost) everything that you can do with SISPPEO as a package can be done using the CLI.
The CLI can also be used to:
print available algorithms and masks (using the algorithms and masks commands)
set up your own workspace, where to store custom algorithms and configuration files without having to add it to SISPPEO and reinstall it
check if each algorithm (included custom ones) is correctly registered (i.e., no missing parts) (using the check_registration command).