def test_encode_features_matches_calculate_feature_matrix():
df = pd.DataFrame({"category": ["b", "c", "d", "e"]}).astype(
{"category": "category"}
)
pd_es = EntitySet("test")
pd_es.add_dataframe(
dataframe_name="a", dataframe=df, index="index", make_index=True
)
features, feature_defs = dfs(entityset=pd_es, target_dataframe_name="a")
features_enc, feature_defs_enc = encode_features(
features, feature_defs, to_encode=["category"]
)
features_calc = calculate_feature_matrix(feature_defs_enc, entityset=pd_es)
pd.testing.assert_frame_equal(features_enc, features_calc)
assert features_calc.ww._schema == features_enc.ww._schema
def test_encode_features_drop_first():
df = pd.DataFrame({"category": ["ao", "b", "c", "d", "e"]}).astype(
{"category": "category"}
)
pd_es = EntitySet("test")
pd_es.add_dataframe(
dataframe_name="a", dataframe=df, index="index", make_index=True
)
features, feature_defs = dfs(entityset=pd_es, target_dataframe_name="a")
features_enc, _ = encode_features(
features, feature_defs, drop_first=True, include_unknown=False
)
assert len(features_enc.columns) == 4
features_enc, feature_defs = encode_features(
features, feature_defs, top_n=3, drop_first=True, include_unknown=False
)
assert len(features_enc.columns) == 2
def test_encode_features_matches_calculate_feature_matrix():
df = pd.DataFrame({
'category': ['b', 'c', 'd', 'e']
}).astype({'category': 'category'})
pd_es = EntitySet('test')
pd_es.add_dataframe(dataframe_name='a',
dataframe=df,
index='index',
make_index=True)
features, feature_defs = dfs(entityset=pd_es, target_dataframe_name='a')
features_enc, feature_defs_enc = encode_features(features,
feature_defs,
to_encode=['category'])
features_calc = calculate_feature_matrix(feature_defs_enc, entityset=pd_es)
pd.testing.assert_frame_equal(features_enc, features_calc)
assert features_calc.ww._schema == features_enc.ww._schema
def test_encode_unknown_features():
# Dataframe with categorical column with "unknown" string
df = pd.DataFrame({
'category': ['unknown', 'b', 'c', 'd', 'e']
}).astype({'category': 'category'})
pd_es = EntitySet('test')
pd_es.add_dataframe(dataframe_name='a',
dataframe=df,
index='index',
make_index=True)
features, feature_defs = dfs(entityset=pd_es, target_dataframe_name='a')
# Specify unknown token for replacement
features_enc, _ = encode_features(features,
feature_defs,
include_unknown=True)
assert list(features_enc.columns) == [
'category = unknown', 'category = e', 'category = d', 'category = c',
'category = b', 'category is unknown'
]
def test_encode_unknown_features():
# Dataframe with categorical column with "unknown" string
df = pd.DataFrame({"category": ["unknown", "b", "c", "d", "e"]}).astype(
{"category": "category"}
)
pd_es = EntitySet("test")
pd_es.add_dataframe(
dataframe_name="a", dataframe=df, index="index", make_index=True
)
features, feature_defs = dfs(entityset=pd_es, target_dataframe_name="a")
# Specify unknown token for replacement
features_enc, _ = encode_features(features, feature_defs, include_unknown=True)
assert list(features_enc.columns) == [
"category = unknown",
"category = e",
"category = d",
"category = c",
"category = b",
"category is unknown",
]
def test_encode_features_drop_first():
df = pd.DataFrame({
'category': ['ao', 'b', 'c', 'd', 'e']
}).astype({'category': 'category'})
pd_es = EntitySet('test')
pd_es.add_dataframe(dataframe_name='a',
dataframe=df,
index='index',
make_index=True)
features, feature_defs = dfs(entityset=pd_es, target_dataframe_name='a')
features_enc, _ = encode_features(features,
feature_defs,
drop_first=True,
include_unknown=False)
assert len(features_enc.columns) == 4
features_enc, feature_defs = encode_features(features,
feature_defs,
top_n=3,
drop_first=True,
include_unknown=False)
assert len(features_enc.columns) == 2