Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Propagate DataArray and Dataset level attributes in coarsening functions and regrid_vertical #513

Merged
merged 7 commits into from
Jul 29, 2020
Merged

Propagate DataArray and Dataset level attributes in coarsening functions and regrid_vertical #513

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
a1e000f
Add keep_attrs argument to coarsening functions
spencerkclark Jul 29, 2020
861e09c
mypy
spencerkclark Jul 29, 2020
b96fb7f
Add keep_attrs argument to regridz
spencerkclark Jul 29, 2020
3c29982
Remove keep_attrs flag; keep_attrs by default
spencerkclark Jul 29, 2020
f23063b
Remove keep_attrs flag from regrid_vertical
spencerkclark Jul 29, 2020
83e7dd4
Improve typing; reformat and lint
spencerkclark Jul 29, 2020
aba77d7
Rename TypeVar with more conventional style
spencerkclark Jul 29, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
51 changes: 35 additions & 16 deletions external/vcm/tests/test_cubedsphere.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -191,18 +191,28 @@ def test_add_coordinates(coarsened_object_type):
def test_weighted_block_average(object_type):
coarsening_factor = 2
dims = ["x", "y"]
attrs = {"units": "m"}
ds_attrs = {"test": "a"}
data = xr.DataArray(
np.array([[2.0, 6.0], [6.0, 2.0]]), dims=dims, coords=None, name="foo"
np.array([[2.0, 6.0], [6.0, 2.0]]),
dims=dims,
coords=None,
name="foo",
attrs=attrs,
)

if object_type == "Dataset":
data = data.to_dataset()
data.attrs = ds_attrs

weights = xr.DataArray(np.array([[6.0, 2.0], [2.0, 6.0]]), dims=dims, coords=None)

expected = xr.DataArray(np.array([[3.0]]), dims=dims, coords=None, name="foo")
expected = xr.DataArray(
np.array([[3.0]]), dims=dims, coords=None, name="foo", attrs=attrs
)
if object_type == "Dataset":
expected = expected.to_dataset()
expected.attrs = ds_attrs

result = weighted_block_average(
data, weights, coarsening_factor, x_dim="x", y_dim="y"
Expand All @@ -220,10 +230,11 @@ def test_weighted_block_average(object_type):
)
def test_edge_weighted_block_average(data, spacing, factor, edge, expected_data):
dims = ["x_dim", "y_dim"]
da = xr.DataArray(data, dims=dims, coords=None)
attrs = {"units": "m"}
da = xr.DataArray(data, dims=dims, coords=None, attrs=attrs)
weights = xr.DataArray(spacing, dims=dims, coords=None)

expected = xr.DataArray(expected_data, dims=dims, coords=None)
expected = xr.DataArray(expected_data, dims=dims, coords=None, attrs=attrs)

result = edge_weighted_block_average(
da, weights, factor, x_dim="x_dim", y_dim="y_dim", edge=edge
Expand All @@ -236,13 +247,13 @@ def input_dataarray():
shape = (4, 4, 2)
data = np.arange(np.product(shape)).reshape(shape).astype(np.float32)
dims = ["x", "y", "z"]
return xr.DataArray(data, dims=dims, coords=None, name="foo")
return xr.DataArray(data, dims=dims, coords=None, name="foo", attrs={"units": "m"})


@pytest.fixture()
def input_dataset(input_dataarray):
bar = xr.DataArray([1, 2, 3], dims=["t"], name="bar")
return xr.merge([input_dataarray, bar])
bar = xr.DataArray([1, 2, 3], dims=["t"], name="bar", attrs={"units": "m"})
return xr.merge([input_dataarray, bar]).assign_attrs(test="a")


@pytest.mark.parametrize("reduction_function", [np.mean, np.median])
Expand All @@ -253,7 +264,11 @@ def test_xarray_block_reduce_dataarray(reduction_function, use_dask, input_dataa
input_dataarray.values, block_size=block_size, func=reduction_function
)
expected = xr.DataArray(
expected_data, dims=input_dataarray.dims, coords=None, name="foo"
expected_data,
dims=input_dataarray.dims,
coords=None,
name="foo",
attrs={"units": "m"},
)

if use_dask:
Expand Down Expand Up @@ -299,6 +314,7 @@ def test_block_reduce_dataarray_coordinates(input_dataarray, coord_func):
input_dataarray.coarsen(x=2, y=2, coord_func=coord_func)
.median()
.rename(input_dataarray.name)
.assign_attrs({"units": "m"})
)
assert_identical_including_dtype(result, expected)

Expand All @@ -323,7 +339,7 @@ def test_horizontal_block_reduce_dataset(input_dataset):

# No change expected to bar, because it contains no horizontal dimensions.
expected_bar = input_dataset.bar
expected = xr.merge([expected_foo, expected_bar])
expected = xr.merge([expected_foo, expected_bar]).assign_attrs(test="a")

result = horizontal_block_reduce(
input_dataset, coarsening_factor, np.median, "x", "y"
Expand Down Expand Up @@ -368,8 +384,9 @@ def test_block_coarsen(input_dataarray):
)
def test_block_edge_sum(data, factor, edge, expected_data):
dims = ["x_dim", "y_dim"]
da = xr.DataArray(data, dims=dims, coords=None)
expected = xr.DataArray(expected_data, dims=dims, coords=None)
attrs = {"units": "m"}
da = xr.DataArray(data, dims=dims, coords=None, attrs=attrs)
expected = xr.DataArray(expected_data, dims=dims, coords=None, attrs=attrs)
result = block_edge_sum(da, factor, x_dim="x_dim", y_dim="y_dim", edge=edge)
assert_identical_including_dtype(result, expected)

Expand Down Expand Up @@ -692,10 +709,11 @@ def test__block_mode():
[1.0, 1.0, 0.0, np.nan],
]
)
da = xr.DataArray(data, dims=["x", "y"])
attrs = {"units": "m"}
da = xr.DataArray(data, dims=["x", "y"], attrs=attrs)

expected_data = np.array([[0.0, 1.0], [1.0, 0.0]])
expected = xr.DataArray(expected_data, dims=["x", "y"])
expected = xr.DataArray(expected_data, dims=["x", "y"], attrs=attrs)

result = _block_mode(da, 2, x_dim="x", y_dim="y", nan_policy="omit")
assert_identical_including_dtype(result, expected)
Expand All @@ -710,13 +728,14 @@ def test_block_mode_via_block_coarsen():
[1.0, 1.0, 0.0, np.nan],
]
)
da = xr.DataArray(data, dims=["x", "y"])
attrs = {"units": "m"}
da = xr.DataArray(data, dims=["x", "y"], attrs=attrs)

expected_data = np.array([[0.0, 1.0], [1.0, 0.0]])
expected = xr.DataArray(expected_data, dims=["x", "y"])
expected = xr.DataArray(expected_data, dims=["x", "y"], attrs=attrs)

result = block_coarsen(
da, 2, x_dim="x", y_dim="y", method="mode", func_kwargs={"nan_policy": "omit"}
da, 2, x_dim="x", y_dim="y", method="mode", func_kwargs={"nan_policy": "omit"},
)
assert_identical_including_dtype(result, expected)

Expand Down
12 changes: 12 additions & 0 deletions external/vcm/tests/test_regridz.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -140,6 +140,18 @@ def test_regrid_vertical_invalid_vertical_dimension_size():
)


@pytest.mark.skipif(not has_mappm, reason="test requires mappm")
def test_regrid_vertical_keep_attrs():
attrs = {"units": "m"}
p_in = input_dataarray((4, 6), z_dim_name="z_outer")
f_in = input_dataarray((4, 5), z_dim_name="z_center").assign_attrs(attrs)
p_out = input_dataarray((4, 3), z_dim_name="z_outer")
f_out = regrid_vertical(
p_in, f_in, p_out, z_dim_center="z_center", z_dim_outer="z_outer",
)
assert f_out.attrs == attrs


def test__mask_weights():
weights = input_dataarray((2, 2, 1)).isel(z=0)
phalf_coarse_on_fine = input_dataarray((2, 2, 3))
Expand Down
Loading