def value_filter(self, feature, values):
"""
Filters the data set based on its values for a given feature.
Args:
feature: string
The name of the feature whose value will be examined for each
sample.
values: single value or list of values.
Samples passing through the filter must have one of these
values for the specified feature.
Returns:
filtered: model.DataSet
The filtered data set.
"""
samples = pandas_util.find(self.get_column(feature), values)
return self.sample_filter(samples)
def value_filter(self, feature, values):
"""
Filters the data set based on its values for a given feature.
Args:
feature: string
The name of the feature whose value will be examined for each
sample.
values: single value or list of values.
Samples passing through the filter must have one of these
values for the specified feature.
Returns:
filtered: model.DataSet
The filtered data set.
"""
samples = pandas_util.find(self.get_column(feature), values)
return self.sample_filter(samples)
def label_filter(self, labels):
"""
Filters the data set based on its labels.
Args:
labels: single value or list of values
Samples with one of these labels will remain in the filtered data
set. All others will be removed.
Returns:
filtered: model.DataSet
The filtered data set.
Raises:
UnlabelledDataSetError if the data set is not labeled.
"""
if not self.is_labelled():
raise UnlabelledDataSetError()
return self.sample_filter(pandas_util.find(self.labels, labels))
def label_filter(self, labels):
"""
Filters the data set based on its labels.
Args:
labels: single value or list of values
Samples with one of these labels will remain in the filtered data
set. All others will be removed.
Returns:
filtered: model.DataSet
The filtered data set.
Raises:
UnlabelledDataSetError if the data set is not labeled.
"""
if not self.is_labelled():
raise UnlabelledDataSetError()
return self.sample_filter(pandas_util.find(self.labels, labels))
def test_find_multiple_values(self):
series = pd.Series(["hostile", "friendly", "friendly", "not_friendly"],
index=["wolf", "cat", "dog", "mouse"])
indices = pandas_util.find(series, ["friendly", "not_friendly"])
assert_that(indices, contains("cat", "dog", "mouse"))
def test_find_one_value(self):
series = pd.Series(["friendly", "friendly", "not_friendly"],
index=["cat", "dog", "mouse"])
indices = pandas_util.find(series, "friendly")
assert_that(indices, contains("cat", "dog"))
def test_find_multiple_values(self):
series = pd.Series(["hostile", "friendly", "friendly", "not_friendly"],
index=["wolf", "cat", "dog", "mouse"])
indices = pandas_util.find(series, ["friendly", "not_friendly"])
assert_that(indices, contains("cat", "dog", "mouse"))
def test_find_one_value(self):
series = pd.Series(["friendly", "friendly", "not_friendly"],
index=["cat", "dog", "mouse"])
indices = pandas_util.find(series, "friendly")
assert_that(indices, contains("cat", "dog"))
