def test_run_udf_code_apply_timeseries_tb():
udf_code = textwrap.dedent("""
import pandas as pd
def apply_timeseries(series: pd.Series, context: dict) -> pd.Series:
return series - series.mean()
""")
xdc = _build_xdc(ts=[2018, 2019, 2020, 2021],
bands=["red", "green", "blue"])
udf_data = UdfData(datacube_list=[xdc])
result = run_udf_code(code=udf_code, data=udf_data)
aa, = result.get_datacube_list()
assert isinstance(aa, XarrayDataCube)
expected_dims = [
{
'name': 't',
'coordinates': [2018, 2019, 2020, 2021]
},
{
'name': 'bands',
'coordinates': ["red", "green", "blue"]
},
]
assert aa.to_dict() == {
'data': [[-15, -15, -15], [-5, -5, -5], [5, 5, 5], [15, 15, 15]],
'dimensions': expected_dims,
}
def test_run_udf_code_reduce_time_mean():
udf_code = _get_udf_code("reduce_time_mean.py")
a = _build_txy_data(ts=[2018, 2019, 2020, 2021],
xs=[2, 3],
ys=[10, 20, 30],
name="temp",
offset=2)
b = _build_txy_data(ts=[2018, 2019, 2020, 2021],
xs=[2, 3],
ys=[10, 20, 30],
name="prec",
offset=4)
udf_data = UdfData(datacube_list=[a, b])
result = run_udf_code(code=udf_code, data=udf_data)
aa, bb = result.get_datacube_list()
assert aa.id == "temp_mean"
assert aa.array.name == "temp_mean"
assert aa.array.shape == (2, 3)
assert list(
aa.array.values.ravel()) == [152.0, 153.0, 154.0, 162.0, 163.0, 164.0]
assert bb.id == "prec_mean"
assert bb.array.name == "prec_mean"
assert bb.array.shape == (2, 3)
assert list(
bb.array.values.ravel()) == [154.0, 155.0, 156.0, 164.0, 165.0, 166.0]
def test_run_udf_code_statistics():
udf_code = _get_udf_code("statistics.py")
xdc = _build_txy_data(ts=[2018, 2019],
xs=[0, 1],
ys=[0, 1, 2],
name="temp",
offset=2)
udf_data = UdfData(datacube_list=[xdc])
result = run_udf_code(code=udf_code, data=udf_data)
structured, = result.structured_data_list
assert structured.to_dict() == {
"data": {
"temp": {
"min": 2,
"mean": 58,
"max": 114,
"sum": 696
}
},
"type":
"dict",
"description":
"Statistical data sum, min, max and mean for each raster collection cube as dict"
}
def test_run_udf_code_ndvi():
udf_code = _get_udf_code("ndvi02.py")
red = _build_txy_data(ts=[2018],
xs=[0, 1],
ys=[0, 1, 2],
name="red",
offset=2)
nir = _build_txy_data(ts=[2018],
xs=[0, 1],
ys=[0, 1, 2],
name="nir",
offset=4)
udf_data = UdfData(datacube_list=[red, nir])
result = run_udf_code(code=udf_code, data=udf_data)
print(nir.array)
print(red.array)
n, = result.get_datacube_list()
assert n.id == "NDVI"
assert n.array.shape == (1, 2, 3)
def ndvi(red, nir):
return (nir - red) / (nir + red)
assert list(n.array.values.ravel()) == [
ndvi(2, 4),
ndvi(3, 5),
ndvi(4, 6),
ndvi(12, 14),
ndvi(13, 15),
ndvi(14, 16)
]
def test_run_udf_code_apply_timeseries_txy():
udf_code = textwrap.dedent("""
import pandas as pd
def apply_timeseries(series: pd.Series, context: dict) -> pd.Series:
return series - series.mean()
""")
a = _build_txy_data(ts=[2018, 2019, 2020, 2021],
xs=[2, 3],
ys=[10, 20, 30],
name="temp",
t_factor=2)
b = _build_txy_data(ts=[2018, 2019, 2020, 2021],
xs=[2, 3],
ys=[10, 20, 30],
name="prec",
t_factor=10)
udf_data = UdfData(datacube_list=[a, b])
result = run_udf_code(code=udf_code, data=udf_data)
aa, bb = result.get_datacube_list()
assert isinstance(aa, XarrayDataCube)
expected_dims = [{
'name': 't',
'coordinates': [2018, 2019, 2020, 2021]
}, {
'name': 'x',
'coordinates': [2, 3]
}, {
'name': 'y',
'coordinates': [10, 20, 30]
}]
assert aa.to_dict() == {
'id':
'temp',
'data': [
[[-3, -3, -3], [-3, -3, -3]],
[[-1, -1, -1], [-1, -1, -1]],
[[1, 1, 1], [1, 1, 1]],
[[3, 3, 3], [3, 3, 3]],
],
'dimensions':
expected_dims,
}
assert isinstance(bb, XarrayDataCube)
assert bb.to_dict() == {
'id':
'prec',
'data': [[[-15, -15, -15], [-15, -15, -15]],
[[-5, -5, -5], [-5, -5, -5]], [[5, 5, 5], [5, 5, 5]],
[[15, 15, 15], [15, 15, 15]]],
'dimensions':
expected_dims,
}
def test_run_udf_code_map_fabs():
udf_code = _get_udf_code("map_fabs.py")
xdc = _build_txy_data(ts=[2019, 2020, 2021],
xs=[2, 3, 4],
ys=[10, 20, 30],
name="temp",
x_factor=-10)
udf_data = UdfData(datacube_list=[xdc])
result = run_udf_code(code=udf_code, data=udf_data)
assert list(xdc.array[0, :2, :].values.ravel()) == [0, 1, 2, -10, -9, -8]
output, = result.get_datacube_list()
assert output.id == "temp_fabs"
assert output.array.name == "temp_fabs"
assert output.array.shape == (3, 3, 3)
assert list(output.array[0, :2, :].values.ravel()) == [0, 1, 2, 10, 9, 8]
def test_run_udf_code_reduce_time_min_median_max():
udf_code = _get_udf_code("reduce_time_min_median_max.py")
a = _build_txy_data(ts=[2018, 2019, 2020],
xs=[2, 3],
ys=[10, 20, 30],
name="temp",
offset=2)
udf_data = UdfData(datacube_list=[a])
result = run_udf_code(code=udf_code, data=udf_data)
x, y, z = result.get_datacube_list()
assert x.id == "temp_min"
assert x.array.shape == (2, 3)
assert list(x.array.values.ravel()) == [2, 3, 4, 12, 13, 14]
assert y.id == "temp_median"
assert y.array.shape == (2, 3)
assert list(y.array.values.ravel()) == [102., 103., 104., 112., 113., 114.]
assert z.id == "temp_max"
assert z.array.shape == (2, 3)
assert list(z.array.values.ravel()) == [202., 203., 204., 212., 213., 214.]