Esempio n. 1
0
    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")
Esempio n. 2
0
    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")
Esempio n. 3
0
    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")
Esempio n. 4
0
    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")
Esempio n. 5
0
    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)
Esempio n. 6
0
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
Esempio n. 7
0
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