def test_binarize():
binarize_spec = Preprocessing(name="binarize", kwargs={"threshold": 14})
data = xr.DataArray(np.arange(30).reshape(2, 3, 5), dims=("x", "y", "c"))
expected = xr.zeros_like(data)
expected[{"x": slice(1, None)}] = 1
preprocessing = make_preprocessing([binarize_spec])
result = preprocessing(data)
xr.testing.assert_allclose(expected, result)
def test_scale_linear_no_channel():
spec = Preprocessing(name="scale_linear",
kwargs={
"offset": 1,
"gain": 2,
"axes": "yx"
})
data = xr.DataArray(np.arange(6).reshape(2, 3), dims=("x", "y"))
expected = xr.DataArray(np.array([[1, 3, 5], [7, 9, 11]]), dims=("x", "y"))
preprocessing = make_preprocessing([spec])
result = preprocessing(data)
xr.testing.assert_allclose(expected, result)
def test_scale_linear():
spec = Preprocessing(name="scale_linear",
kwargs={
"offset": [1, 2, 42],
"gain": [1, 2, 3],
"axes": "yx"
})
data = xr.DataArray(np.arange(6).reshape(1, 2, 3), dims=("x", "y", "c"))
expected = xr.DataArray(np.array([[[1, 4, 48], [4, 10, 57]]]),
dims=("x", "y", "c"))
preprocessing = make_preprocessing([spec])
result = preprocessing(data)
xr.testing.assert_allclose(expected, result)
def test_zero_mean_unit_variance_preprocessing():
zero_mean_spec = Preprocessing(name="zero_mean_unit_variance", kwargs={})
data = xr.DataArray(np.arange(9).reshape(3, 3), dims=("x", "y"))
expected = xr.DataArray(
np.array([
[-1.54919274, -1.16189455, -0.77459637],
[-0.38729818, 0.0, 0.38729818],
[0.77459637, 1.16189455, 1.54919274],
]),
dims=("x", "y"),
)
preprocessing = make_preprocessing([zero_mean_spec])
result = preprocessing(data)
xr.testing.assert_allclose(expected, result)
def test_combination_of_preprocessing_steps_with_dims_specified():
zero_mean_spec = Preprocessing(name="zero_mean_unit_variance",
kwargs={"axes": ("x", "y")})
data = xr.DataArray(np.arange(18).reshape(2, 3, 3), dims=("c", "x", "y"))
expected = xr.DataArray(
np.array([
[-1.54919274, -1.16189455, -0.77459637],
[-0.38729818, 0.0, 0.38729818],
[0.77459637, 1.16189455, 1.54919274],
]),
dims=("x", "y"),
)
preprocessing = make_preprocessing([ADD_BATCH_DIM, zero_mean_spec])
result = preprocessing(data)
xr.testing.assert_allclose(expected, result[0][0])
def make_ensure_dtype_preprocessing(dtype):
return Preprocessing(name="__tiktorch_ensure_dtype",
kwargs={"dtype": dtype})
from typing import List
import numpy as np
import xarray as xr
from bioimageio.spec.nodes import Preprocessing
from ._types import Transform
ADD_BATCH_DIM = Preprocessing(name="__tiktorch_add_batch_dim", kwargs=None)
def make_ensure_dtype_preprocessing(dtype):
return Preprocessing(name="__tiktorch_ensure_dtype",
kwargs={"dtype": dtype})
def scale_linear(tensor: xr.DataArray, *, gain, offset, axes) -> xr.DataArray:
"""scale the tensor with a fixed multiplicative and additive factor"""
scale_axes = tuple(ax for ax in tensor.dims
if (ax not in axes and ax != "b"))
if scale_axes:
gain = xr.DataArray(np.atleast_1d(gain), dims=scale_axes)
offset = xr.DataArray(np.atleast_1d(offset), dims=scale_axes)
return tensor * gain + offset
def zero_mean_unit_variance(tensor: xr.DataArray,
axes=None,
eps=1.0e-6,
mode="per_sample") -> xr.DataArray:
if axes:
def test_unknown_preprocessing_should_raise():
mypreprocessing = Preprocessing(name="mycoolpreprocessing",
kwargs={"axes": ("x", "y")})
with pytest.raises(NotImplementedError):
make_preprocessing([mypreprocessing])