###############################################################################
# dB transform
# ~~~~~~~~~~~~
# We transform the rainfall data into dB units: 10*log(R)
R_, _ = transformation.dB_transform(R, metadata)
data.append((R_ - np.mean(R_)) / np.std(R_))
labels.append("dB")
skw.append(skew(R_))
###############################################################################
# Square-root transform
# ~~~~~~~~~~~~~~~~~~~~~
# Transform the data using the square-root: sqrt(R)
R_, _ = transformation.sqrt_transform(R, metadata)
data.append((R_ - np.mean(R_)) / np.std(R_))
labels.append("sqrt")
skw.append(skew(R_))
###############################################################################
# Box-Cox transform
# ~~~~~~~~~~~~~~~~~
# We now apply the Box-Cox transform using the best parameter lambda found above.
R_, _ = transformation.boxcox_transform(R, metadata, Lambda)
data.append((R_ - np.mean(R_)) / np.std(R_))
labels.append("Box-Cox\n($\lambda=$%.2f)" % Lambda)
skw.append(skew(R_))
###############################################################################
def test_sqrt_transform(R, metadata, inverse, expected):
"""Test the sqrt_transform."""
assert_array_almost_equal(
transformation.sqrt_transform(R, metadata, inverse)[0], expected
)
