def test_binning_pctile():
binner = BinningTransformer(cols=["a"],
n_bins=3,
strategy="percentile",
return_bin_label=True,
overwrite=False)
binner.fit(iris)
trans = binner.transform(iris)
unq = np.unique(trans["a_binned"].values).tolist()
assert unq == ["(-Inf, 5.40]", "(5.40, 6.30]", "(6.30, Inf]"], unq
def test_binning_complex():
# Test with complex n_bins
binner = BinningTransformer(cols=["a", "b"],
n_bins=[2, 3],
strategy="uniform",
return_bin_label=False,
overwrite=True)
binner.fit(iris)
trans = binner.transform(iris)
# show the columns stayed the same
assert trans.columns.tolist() == iris.columns.tolist()
# assert the different levels of integers
assert_array_equal(np.unique(trans.a.values), [0, 1])
assert_array_equal(np.unique(trans.b.values), [0, 1, 2])
# show both types are now int
assert trans.dtypes['a'].name.startswith("int")
assert trans.dtypes['b'].name.startswith("int")
# Test with overwrite = False
binner.overwrite = False
trans2 = binner.transform(iris)
assert trans2.shape[1] == 6
assert trans2.columns.tolist() == [
"a", "b", "c", "d", "a_binned", "b_binned"
], trans2.columns
def test_binning_simple():
binner = BinningTransformer(cols=["a"],
n_bins=3,
strategy="uniform",
return_bin_label=True,
overwrite=True)
binner.fit(iris)
trans = binner.transform(iris)
# show the dfs are not the same
assert trans is not iris
# show the columns stayed the same, though
assert trans.columns.tolist() == iris.columns.tolist()
# show we have a string datatype now
assert trans.dtypes['a'].name == 'object'
# if we set the return_bin_label to false and then transform again
# show we actually get an integer back
binner.return_bin_label = False
trans2 = binner.transform(iris)
assert trans2.dtypes['a'].name.startswith("int")
# show there are three levels
assert_array_equal(np.unique(trans2.a.values), [0, 1, 2])
def test_binning_corners():
# assertion function to assert fails
def f(binner, exc):
assert_raises(exc, binner.fit, iris)
# this one will fail since n_bins contains a non-specified column
f(BinningTransformer(cols=["a"], n_bins={"b": 2}), ValueError)
# this one will fail for the same reason
f(BinningTransformer(cols=["a", "c"], n_bins={"a": 3, "b": 2}), ValueError)
# this one will fail for a bad integer
f(BinningTransformer(cols=["a", "c"], n_bins=[2, 1]), ValueError)
# this one will fail for a dim mismatch
f(BinningTransformer(cols=["a", "c"], n_bins=[2]), ValueError)
# this one will fail since n_bins is illegal
f(BinningTransformer(cols=["a", "c"], n_bins=None), TypeError)
# this one will fail since strategy is illegal
f(BinningTransformer(cols=["a"], n_bins=3, strategy="illegal"), ValueError)
<br/>
"""
print(__doc__)
# Author: Taylor Smith <*****@*****.**>
from matplotlib import pyplot as plt
from skoot.datasets import load_iris_df
from skoot.preprocessing import BinningTransformer
# #############################################################################
# load data
iris = load_iris_df(include_tgt=False, names=["a", "b", "c", "d"])
binner = BinningTransformer(cols=["a", "b"], return_bin_label=True,
strategy="uniform", overwrite=False,
n_bins=4)
# print the head of the binned dataset
print(binner.fit_transform(iris).head())
# #############################################################################
# Show where the boundaries reside
a_lower = binner.bins_["a"].lower_bounds[1:] # skip the -np.inf
plt.hist(iris["a"].values)
# plot vertical lines where bins are
for bound in a_lower:
plt.axvline(bound, ls="--")
plt.title("Iris feature 'a' + bin markers")
def test_binning_persistable():
assert_persistable(BinningTransformer(), "location.pkl", iris)
def test_binning_asdf():
assert_transformer_asdf(BinningTransformer(), iris)