def test_feature_assembler_invalid_dtype_input_fail():
df = util.iris(['sepallength', 'sepalwidth', 'petalwidth', 'petallength'],
size=10)
test_out = {
'Feat': [[3.1, 0.2], [3.5, 0.3], [3.3, 0.4], [3.1, 0.2], [3.6, 0.2],
[3.9, 0.4], [3.4, 0.3], [3.4, 0.2], [2.9, 0.2], [3.1, 0.1]]
}
test_out_2 = {
'Feat2':
['3.1', '0.3', '0.4', '0.2', '3.6', '3.9', '0.3', '0.2', '2.9', '0.1']
}
test_out = pd.DataFrame(test_out)
test_out_2 = pd.DataFrame(test_out_2)
df = pd.concat([df, test_out, test_out_2], axis=1, join='inner')
arguments = {
'parameters': {
'attributes': ['sepalwidth', 'Feat', 'Feat2'],
'multiplicity': {
'input data': 0
}
},
'named_inputs': {
'input data': 'df',
},
'named_outputs': {
'output data': 'out'
}
}
instance = FeatureAssemblerOperation(**arguments)
with pytest.raises(ValueError) as val_err:
util.execute(instance.generate_code(), {'df': df})
assert "Input 'df' must contain numeric values only for task" in str(
val_err.value)
def test_feature_assembler_alias_param_success():
df = util.iris(['sepallength', 'sepalwidth', 'petalwidth', 'petallength'],
size=10)
test_df = df.copy()
arguments = {
'parameters': {
'attributes': ['sepalwidth', 'petalwidth'],
'multiplicity': {
'input data': 0
},
'alias': 'Feat'
},
'named_inputs': {
'input data': 'df',
},
'named_outputs': {
'output data': 'out'
}
}
instance = FeatureAssemblerOperation(**arguments)
result = util.execute(instance.generate_code(), {'df': df})
test_out = {
'Feat': [[3.5, 0.2], [3.0, 0.2], [3.2, 0.2], [3.1, 0.2], [3.6, 0.2],
[3.9, 0.4], [3.4, 0.3], [3.4, 0.2], [2.9, 0.2], [3.1, 0.1]]
}
test_out = pd.DataFrame(test_out)
test_out = pd.concat([test_df, test_out], axis=1)
assert result['out'].equals(test_out)
def test_feature_assembler_operation_success():
params = {
FeatureAssemblerOperation.ATTRIBUTES_PARAM: ['col'],
}
n_in = {'input data': 'input_1'}
n_out = {'output data': 'output_1'}
in1 = n_in['input data']
out = n_out['output data']
instance = FeatureAssemblerOperation(params,
named_inputs=n_in,
named_outputs=n_out)
code = instance.generate_code()
expected_code = dedent("""
cols = {cols}
{output} = {input}
{output}['FeatureField'] = {input}[cols].values.tolist()
""".format(cols=params[FeatureAssemblerOperation.ATTRIBUTES_PARAM],
output=out,
input=in1))
result, msg = compare_ast(ast.parse(code), ast.parse(expected_code))
assert result, msg + format_code_comparison(code, expected_code)
def test_feature_assembler_missing_input_implies_no_code_success():
arguments = {
'parameters': {
'attributes': ['sepalwidth', 'petalwidth'],
'multiplicity': {
'input data': 0
}
},
'named_inputs': {},
'named_outputs': {
'output data': 'out'
}
}
instance = FeatureAssemblerOperation(**arguments)
assert instance.generate_code() is None
def test_feature_assembler_operation_failure():
params = {FeatureAssemblerOperation.ALIAS_PARAM: 'c'}
with pytest.raises(ValueError):
n_in = {'input data': 'input_1'}
n_out = {'output data': 'output_1'}
FeatureAssemblerOperation(params,
named_inputs=n_in,
named_outputs=n_out)
def test_feature_assembler_missing_multiplicity_param_fail():
df = util.iris(['sepallength', 'sepalwidth', 'petalwidth', 'petallength'],
size=10)
arguments = {
'parameters': {
'attributes': ['sepalwidth', 'petalwidth']
},
'named_inputs': {
'input data': 'df',
},
'named_outputs': {
'output data': 'out'
}
}
instance = FeatureAssemblerOperation(**arguments)
with pytest.raises(KeyError) as key_err:
util.execute(instance.generate_code(), {'df': df})
assert "'multiplicity'" in str(key_err.value)
def test_feature_assembler_invalid_multiplicity_param_fail():
df = util.iris(['sepallength', 'sepalwidth', 'petalwidth', 'petallength'],
size=10)
arguments = {
'parameters': {
'attributes': ['sepalwidth', 'petalwidth'],
'multiplicity': 'invalid'
},
'named_inputs': {
'input data': 'df',
},
'named_outputs': {
'output data': 'out'
}
}
instance = FeatureAssemblerOperation(**arguments)
with pytest.raises(TypeError) as typ_err:
util.execute(instance.generate_code(), {'df': df})
assert "string indices must be integers" in str(typ_err)
def test_feature_assembler_invalid_attributes_param_fail():
df = util.iris(['sepallength', 'sepalwidth', 'petalwidth', 'petallength'],
size=10)
arguments = {
'parameters': {
'attributes': 'invalid',
'multiplicity': {
'input data': 0
}
},
'named_inputs': {
'input data': 'df',
},
'named_outputs': {
'output data': 'out'
}
}
instance = FeatureAssemblerOperation(**arguments)
with pytest.raises(NameError) as nam_err:
util.execute(instance.generate_code(), {'df': df})
assert "name 'invalid' is not defined" in str(nam_err)
def test_feature_assembler_na_drop_success():
"""
The chained assignment warnings / exceptions are aiming to inform the user
of a possibly invalid assignment. There may be false positives; situations
where a chained assignment is inadvertently reported.
https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
"""
df = util.iris(['sepallength', 'sepalwidth', 'petalwidth', 'petallength'],
size=10)
# Line below is causing a false positive warning (?)
df.loc[0:2, 'sepalwidth'] = np.NaN
test_df = df.copy()
arguments = {
'parameters': {
'attributes': ['sepalwidth', 'petalwidth'],
'multiplicity': {
'input data': 0
}
},
'named_inputs': {
'input data': 'df',
},
'named_outputs': {
'output data': 'out'
}
}
instance = FeatureAssemblerOperation(**arguments)
result = util.execute(instance.generate_code(), {'df': df})
test_out = {
'FeatureField': [
np.NaN, np.NaN, np.NaN, [3.1, 0.2], [3.6, 0.2], [3.9, 0.4],
[3.4, 0.3], [3.4, 0.2], [2.9, 0.2], [3.1, 0.1]
]
}
test_out = pd.DataFrame(test_out)
test_out = pd.concat([test_df, test_out], axis=1)
test_out.drop([0, 1, 2], axis=0, inplace=True)
assert result['out'].equals(test_out)
def test_feature_assembler_missing_attributes_param_fail():
arguments = {
'parameters': {
'multiplicity': {
'input data': 0
}
},
'named_inputs': {
'input data': 'df',
},
'named_outputs': {
'output data': 'out'
}
}
with pytest.raises(ValueError) as val_err:
FeatureAssemblerOperation(**arguments)
assert "Parameters 'attributes' must be informed for task" in str(
val_err.value)