def test_fornav_swath_smaller(self):
"""Test that a swath smaller than the output grid is entirely used."""
from pyresample.ewa import _fornav
swath_shape = (1600, 3200)
data_type = np.float32
# Create a fake row and cols array
rows = np.empty(swath_shape, dtype=np.float32)
rows[:] = np.linspace(500, 800, 1600)[:, None]
cols = np.empty(swath_shape, dtype=np.float32)
cols[:] = np.linspace(200, 600, 3200)
rows_per_scan = 16
# Create a fake data swath
data = np.ones(swath_shape, dtype=data_type)
out = np.empty((1000, 1000), dtype=data_type)
grid_points_covered = _fornav.fornav_wrapper(cols, rows, (data, ),
(out, ), np.nan, np.nan,
rows_per_scan)
one_grid_points_covered = grid_points_covered[0]
# The swath was smaller than the grid, make sure its whole area
# was covered (percentage of grid rows/cols to overall size)
self.assertAlmostEqual(one_grid_points_covered / float(out.size),
0.12,
2,
msg="Not all input swath pixels were used")
# The swath was all 1s so there shouldn't be any non-1 values in the
# output except outside the swath
self.assertTrue(((out == 1) | np.isnan(out)).all(),
msg="Unexpected interpolation values were returned")
def test_fornav_swath_larger(self):
"""Test that a swath larger than the output grid fills the entire grid."""
from pyresample.ewa import _fornav
swath_shape = (1600, 3200)
data_type = np.float32
# Create a fake row and cols array
rows = np.empty(swath_shape, dtype=np.float32)
rows[:] = np.linspace(-500, 2500, 1600)[:, None]
cols = np.empty(swath_shape, dtype=np.float32)
cols[:] = np.linspace(-2500, 1500, 3200)
rows_per_scan = 16
# Create a fake data swath
data = np.ones(swath_shape, dtype=data_type)
out = np.empty((1000, 1000), dtype=data_type)
grid_points_covered = _fornav.fornav_wrapper(cols, rows, (data, ),
(out, ), np.nan, np.nan,
rows_per_scan)
one_grid_points_covered = grid_points_covered[0]
# The swath was larger than the grid, all of the grid should have
# been covered by swath pixels
self.assertEqual(one_grid_points_covered,
out.size,
msg="Not all grid pixels were filled")
# The swath was all 1s so there shouldn't be any non-1 values in the
# output except outside the swath
self.assertTrue(((out == 1) | np.isnan(out)).all(),
msg="Unexpected interpolation values were returned")
def test_fornav_swath_smaller_int8(self):
"""Test that a swath smaller than the output grid is entirely used.
"""
from pyresample.ewa import _fornav
swath_shape = (1600, 3200)
data_type = np.int8
# Create a fake row and cols array
rows = np.empty(swath_shape, dtype=np.float32)
rows[:] = np.linspace(500, 800, 1600)[:, None]
cols = np.empty(swath_shape, dtype=np.float32)
cols[:] = np.linspace(200, 600, 3200)
rows_per_scan = 16
# Create a fake data swath
data = np.ones(swath_shape, dtype=data_type)
out = np.empty((1000, 1000), dtype=data_type)
grid_points_covered = _fornav.fornav_wrapper(cols, rows, (data,), (out,),
-128, -128, rows_per_scan)
one_grid_points_covered = grid_points_covered[0]
# The swath was smaller than the grid, make sure its whole area
# was covered (percentage of grid rows/cols to overall size)
self.assertAlmostEqual(one_grid_points_covered / float(out.size), 0.12, 2,
msg="Not all input swath pixels were used")
# The swath was all 1s so there shouldn't be any non-1 values in the
# output except outside the swath
# import ipdb; ipdb.set_trace()
self.assertTrue(((out == 1) | (out == -128)).all(),
msg="Unexpected interpolation values were returned")
def test_fornav_swath_larger(self):
"""Test that a swath larger than the output grid fills the entire grid.
"""
from pyresample.ewa import _fornav
swath_shape = (1600, 3200)
data_type = np.float32
# Create a fake row and cols array
rows = np.empty(swath_shape, dtype=np.float32)
rows[:] = np.linspace(-500, 2500, 1600)[:, None]
cols = np.empty(swath_shape, dtype=np.float32)
cols[:] = np.linspace(-2500, 1500, 3200)
rows_per_scan = 16
# Create a fake data swath
data = np.ones(swath_shape, dtype=data_type)
out = np.empty((1000, 1000), dtype=data_type)
grid_points_covered = _fornav.fornav_wrapper(cols, rows, (data,), (out,),
np.nan, np.nan, rows_per_scan)
one_grid_points_covered = grid_points_covered[0]
# The swath was larger than the grid, all of the grid should have
# been covered by swath pixels
self.assertEqual(one_grid_points_covered, out.size,
msg="Not all grid pixels were filled")
# The swath was all 1s so there shouldn't be any non-1 values in the
# output except outside the swath
self.assertTrue(((out == 1) | np.isnan(out)).all(),
msg="Unexpected interpolation values were returned")
def test_fornav_swath_wide_input(self):
"""Test that a swath with large input pixels on the left edge of the output."""
from pyresample.ewa import _fornav
swath_shape = (400, 800)
data_type = np.float32
# Create a fake row and cols array
rows = np.empty(swath_shape, dtype=np.float32)
rows[:] = np.linspace(-500, 500, 400)[:, None]
cols = np.empty(swath_shape, dtype=np.float32)
cols[:] = np.linspace(-500, 500, 800) + 0.5
rows_per_scan = 16
# Create a fake data swath
data = np.ones(swath_shape, dtype=data_type)
out = np.empty((800, 1000), dtype=data_type)
grid_points_covered = _fornav.fornav_wrapper(cols, rows, (data,), (out,),
np.nan, np.nan, rows_per_scan)
one_grid_points_covered = grid_points_covered[0]
# the upper-left 500x500 square should be filled with 1s at least
assert 500 * 500 <= one_grid_points_covered <= 505 * 505
np.testing.assert_allclose(out[:500, :500], 1)
def fornav(cols, rows, area_def, data_in,
rows_per_scan=None, fill=None, out=None,
weight_count=10000, weight_min=0.01, weight_distance_max=1.0,
weight_delta_max=10.0, weight_sum_min=-1.0,
maximum_weight_mode=False):
"""Remap data in to output grid using elliptical weighted averaging.
This algorithm works under the assumption that the data is observed
one scan line at a time. However, good results can still be achieved
for non-scan based data is provided if `rows_per_scan` is set to the
number of rows in the entire swath or by setting it to `None`.
Parameters
----------
cols : numpy array
Column location for each input swath pixel (from `ll2cr`)
rows : numpy array
Row location for each input swath pixel (from `ll2cr`)
area_def : AreaDefinition
Grid definition to be mapped to
data_in : numpy array or tuple of numpy arrays
Swath data to be remapped to output grid
rows_per_scan : int or None, optional
Number of data rows for every observed scanline. If None then the
entire swath is treated as one large scanline.
fill : float/int or None, optional
If `data_in` is made of numpy arrays then this represents the fill
value used to mark invalid data pixels. This value will also be
used in the output array(s). If None, then np.nan will be used
for float arrays and -999 will be used for integer arrays.
out : numpy array or tuple of numpy arrays, optional
Specify a numpy array to be written to for each input array. This can
be used as an optimization by providing `np.memmap` arrays or other
array-like objects.
weight_count : int, optional
number of elements to create in the gaussian weight table.
Default is 10000. Must be at least 2
weight_min : float, optional
the minimum value to store in the last position of the
weight table. Default is 0.01, which, with a
`weight_distance_max` of 1.0 produces a weight of 0.01
at a grid cell distance of 1.0. Must be greater than 0.
weight_distance_max : float, optional
distance in grid cell units at which to
apply a weight of `weight_min`. Default is
1.0. Must be greater than 0.
weight_delta_max : float, optional
maximum distance in grid cells in each grid
dimension over which to distribute a single swath cell.
Default is 10.0.
weight_sum_min : float, optional
minimum weight sum value. Cells whose weight sums
are less than `weight_sum_min` are set to the grid fill value.
Default is EPSILON.
maximum_weight_mode : bool, optional
If False (default), a weighted average of
all swath cells that map to a particular grid cell is used.
If True, the swath cell having the maximum weight of all
swath cells that map to a particular grid cell is used. This
option should be used for coded/category data, i.e. snow cover.
Returns
-------
(valid grid points, output arrays): tuple of integer tuples and numpy array tuples
The valid_grid_points tuple holds the number of output grid pixels that
were written with valid data. The second element in the tuple is a tuple of
output grid numpy arrays for each input array. If there was only one input
array provided then the returned tuple is simply the singe points integer
and single output grid array.
"""
if isinstance(data_in, (tuple, list)):
# we can only support one data type per call at this time
assert(in_arr.dtype == data_in[0].dtype for in_arr in data_in[1:])
else:
# assume they gave us a single numpy array-like object
data_in = [data_in]
# need a list for replacing these arrays later
data_in = list(data_in)
# determine a fill value if they didn't tell us what they have as a
# fill value in the numpy arrays
if "fill" is None:
if np.issubdtype(data_in[0].dtype, np.floating):
fill = np.nan
elif np.issubdtype(data_in[0].dtype, np.integer):
fill = -999
else:
raise ValueError("Unsupported input data type for EWA Resampling: {}".format(data_in[0].dtype))
convert_to_masked = False
for idx, in_arr in enumerate(data_in):
if isinstance(in_arr, np.ma.MaskedArray):
convert_to_masked = True
# convert masked arrays to single numpy arrays
data_in[idx] = in_arr.filled(fill)
data_in = tuple(data_in)
if out is not None:
# the user may have provided memmapped arrays or other array-like objects
out = tuple("out")
else:
# create a place for output data to be written
out = tuple(np.empty(area_def.shape, dtype=in_arr.dtype) for in_arr in data_in)
# see if the user specified rows per scan
# otherwise, use the entire swath as one "scanline"
rows_per_scan = rows_per_scan or data_in[0].shape[0]
results = _fornav.fornav_wrapper(cols, rows, data_in, out,
np.nan, np.nan, rows_per_scan)
def _mask_helper(data, fill):
if np.isnan(fill):
return np.isnan(data)
else:
return data == fill
if convert_to_masked:
# they gave us masked arrays so give them masked arrays back
out = [np.ma.masked_where(_mask_helper(out_arr, fill), out_arr) for out_arr in out]
if len(out) == 1:
# they only gave us one data array as input, so give them one back
out = out[0]
results = results[0]
return results, out
def fornav(cols,
rows,
area_def,
data_in,
rows_per_scan=None,
fill=None,
out=None,
weight_count=10000,
weight_min=0.01,
weight_distance_max=1.0,
weight_delta_max=10.0,
weight_sum_min=-1.0,
maximum_weight_mode=False):
"""Remap data in to output grid using elliptical weighted averaging.
This algorithm works under the assumption that the data is observed
one scan line at a time. However, good results can still be achieved
for non-scan based data is provided if `rows_per_scan` is set to the
number of rows in the entire swath or by setting it to `None`.
Parameters
----------
cols : numpy array
Column location for each input swath pixel (from `ll2cr`)
rows : numpy array
Row location for each input swath pixel (from `ll2cr`)
area_def : AreaDefinition
Grid definition to be mapped to
data_in : numpy array or tuple of numpy arrays
Swath data to be remapped to output grid
rows_per_scan : int or None, optional
Number of data rows for every observed scanline. If None then the
entire swath is treated as one large scanline.
fill : float/int or None, optional
If `data_in` is made of numpy arrays then this represents the fill
value used to mark invalid data pixels. This value will also be
used in the output array(s). If None, then np.nan will be used
for float arrays and -999 will be used for integer arrays.
out : numpy array or tuple of numpy arrays, optional
Specify a numpy array to be written to for each input array. This can
be used as an optimization by providing `np.memmap` arrays or other
array-like objects.
weight_count : int, optional
number of elements to create in the gaussian weight table.
Default is 10000. Must be at least 2
weight_min : float, optional
the minimum value to store in the last position of the
weight table. Default is 0.01, which, with a
`weight_distance_max` of 1.0 produces a weight of 0.01
at a grid cell distance of 1.0. Must be greater than 0.
weight_distance_max : float, optional
distance in grid cell units at which to
apply a weight of `weight_min`. Default is
1.0. Must be greater than 0.
weight_delta_max : float, optional
maximum distance in grid cells in each grid
dimension over which to distribute a single swath cell.
Default is 10.0.
weight_sum_min : float, optional
minimum weight sum value. Cells whose weight sums
are less than `weight_sum_min` are set to the grid fill value.
Default is EPSILON.
maximum_weight_mode : bool, optional
If False (default), a weighted average of
all swath cells that map to a particular grid cell is used.
If True, the swath cell having the maximum weight of all
swath cells that map to a particular grid cell is used. This
option should be used for coded/category data, i.e. snow cover.
Returns
-------
(valid grid points, output arrays): tuple of integer tuples and numpy array tuples
The valid_grid_points tuple holds the number of output grid pixels that
were written with valid data. The second element in the tuple is a tuple of
output grid numpy arrays for each input array. If there was only one input
array provided then the returned tuple is simply the singe points integer
and single output grid array.
"""
if isinstance(data_in, (tuple, list)):
# we can only support one data type per call at this time
assert (in_arr.dtype == data_in[0].dtype for in_arr in data_in[1:])
else:
# assume they gave us a single numpy array-like object
data_in = [data_in]
# need a list for replacing these arrays later
data_in = list(data_in)
# determine a fill value if they didn't tell us what they have as a
# fill value in the numpy arrays
if "fill" is None:
if np.issubdtype(data_in[0].dtype, np.floating):
fill = np.nan
elif np.issubdtype(data_in[0].dtype, np.integer):
fill = -999
else:
raise ValueError(
"Unsupported input data type for EWA Resampling: {}".format(
data_in[0].dtype))
convert_to_masked = False
for idx, in_arr in enumerate(data_in):
if isinstance(in_arr, np.ma.MaskedArray):
convert_to_masked = True
# convert masked arrays to single numpy arrays
data_in[idx] = in_arr.filled(fill)
data_in = tuple(data_in)
if out is not None:
# the user may have provided memmapped arrays or other array-like objects
out = tuple("out")
else:
# create a place for output data to be written
out = tuple(
np.empty(area_def.shape, dtype=in_arr.dtype) for in_arr in data_in)
# see if the user specified rows per scan
# otherwise, use the entire swath as one "scanline"
rows_per_scan = rows_per_scan or data_in[0].shape[0]
results = _fornav.fornav_wrapper(cols, rows, data_in, out, np.nan, np.nan,
rows_per_scan)
def _mask_helper(data, fill):
if np.isnan(fill):
return np.isnan(data)
else:
return data == fill
if convert_to_masked:
# they gave us masked arrays so give them masked arrays back
out = [
np.ma.masked_where(_mask_helper(out_arr, fill), out_arr)
for out_arr in out
]
if len(out) == 1:
# they only gave us one data array as input, so give them one back
out = out[0]
results = results[0]
return results, out
