def set_output_summary(self, data):
summary = len(data) if data else self.info.NoOutput
detail = format_summary_details(data) if data else ""
self.info.set_output_summary(summary, detail)
def commit(self):
matching_output = self.data
non_matching_output = None
annotated_output = None
self.Error.clear()
if self.data:
domain = self.data.domain
conditions = []
for attr_name, oper_idx, values in self.conditions:
if attr_name in self.AllTypes:
attr_index = attr = None
attr_type = self.AllTypes[attr_name]
operators = self.Operators[attr_name]
else:
attr_index = domain.index(attr_name)
attr = domain[attr_index]
attr_type = vartype(attr)
operators = self.Operators[type(attr)]
opertype, _ = operators[oper_idx]
if attr_type == 0:
filter = data_filter.IsDefined()
elif attr_type in (2, 4): # continuous, time
try:
floats = self._values_to_floats(attr, values)
except ValueError as e:
self.Error.parsing_error(e.args[0])
return
if floats is None:
continue
filter = data_filter.FilterContinuous(
attr_index, opertype, *floats)
elif attr_type == 3: # string
filter = data_filter.FilterString(
attr_index, opertype, *[str(v) for v in values])
else:
if opertype == FilterDiscreteType.IsDefined:
f_values = None
else:
if not values or not values[0]:
continue
values = [attr.values[i - 1] for i in values]
if opertype == FilterDiscreteType.Equal:
f_values = {values[0]}
elif opertype == FilterDiscreteType.NotEqual:
f_values = set(attr.values)
f_values.remove(values[0])
elif opertype == FilterDiscreteType.In:
f_values = set(values)
else:
raise ValueError("invalid operand")
filter = data_filter.FilterDiscrete(attr_index, f_values)
conditions.append(filter)
if conditions:
self.filters = data_filter.Values(conditions)
matching_output = self.filters(self.data)
self.filters.negate = True
non_matching_output = self.filters(self.data)
row_sel = np.in1d(self.data.ids, matching_output.ids)
annotated_output = create_annotated_table(self.data, row_sel)
# if hasattr(self.data, "name"):
# matching_output.name = self.data.name
# non_matching_output.name = self.data.name
purge_attrs = self.purge_attributes
purge_classes = self.purge_classes
if (purge_attrs or purge_classes) and \
not isinstance(self.data, SqlTable):
attr_flags = sum([
Remove.RemoveConstant * purge_attrs,
Remove.RemoveUnusedValues * purge_attrs
])
class_flags = sum([
Remove.RemoveConstant * purge_classes,
Remove.RemoveUnusedValues * purge_classes
])
# same settings used for attributes and meta features
remover = Remove(attr_flags, class_flags, attr_flags)
matching_output = remover(matching_output)
non_matching_output = remover(non_matching_output)
annotated_output = remover(annotated_output)
if matching_output is not None and not len(matching_output):
matching_output = None
if non_matching_output is not None and not len(non_matching_output):
non_matching_output = None
if annotated_output is not None and not len(annotated_output):
annotated_output = None
self.Outputs.matching_data.send(matching_output)
self.Outputs.unmatched_data.send(non_matching_output)
self.Outputs.annotated_data.send(annotated_output)
self.match_desc = report.describe_data_brief(matching_output)
self.nonmatch_desc = report.describe_data_brief(non_matching_output)
summary = matching_output.approx_len() if matching_output else \
self.info.NoOutput
details = format_summary_details(
matching_output) if matching_output else ""
self.info.set_output_summary(summary, details)
def _set_output_summary(self, data: Optional[Table] = None):
if data:
summary, details = len(data), format_summary_details(data)
else:
summary, details = self.info.NoOutput, ""
self.info.set_output_summary(summary, details)
def _set_output_summary(self, output):
summary = len(output) if output else self.info.NoOutput
details = format_summary_details(output) if output else ""
self.info.set_output_summary(summary, details)
def _set_input_summary(self):
summary = len(self.data) if self.data else self.info.NoInput
details = format_summary_details(self.data) if self.data else ""
self.info.set_input_summary(summary, details)
def test_output(self):
# send data and template data
info = self.widget.info
no_input, no_output = "No data on input", "No data on output"
self.send_signal(self.widget.Inputs.data, self.data[::15])
self.send_signal(self.widget.Inputs.template_data, self.disc_data)
output = self.get_output(self.widget.Outputs.transformed_data)
self.assertTableEqual(output, self.disc_data[::15])
self.assertEqual("Input data with 10 instances and 4 features.",
self.widget.input_label.text())
self.assertEqual("Template domain applied.",
self.widget.template_label.text())
self.assertEqual("Output data includes 4 features.",
self.widget.output_label.text())
data_list = [("Data", self.data[::15]),
("Template data", self.disc_data)]
summary, details = "10, 150", format_multiple_summaries(data_list)
self.assertEqual(info._StateInfo__input_summary.brief, summary)
self.assertEqual(info._StateInfo__input_summary.details, details)
summary, details = "10", format_summary_details(output)
self.assertEqual(info._StateInfo__output_summary.brief, summary)
self.assertEqual(info._StateInfo__output_summary.details, details)
# remove template data
self.send_signal(self.widget.Inputs.template_data, None)
output = self.get_output(self.widget.Outputs.transformed_data)
self.assertIsNone(output)
self.assertEqual("Input data with 10 instances and 4 features.",
self.widget.input_label.text())
self.assertEqual("No template data on input.",
self.widget.template_label.text())
self.assertEqual("", self.widget.output_label.text())
data_list = [("Data", self.data[::15]), ("Template data", None)]
summary, details = "10, 0", format_multiple_summaries(data_list)
self.assertEqual(info._StateInfo__input_summary.brief, summary)
self.assertEqual(info._StateInfo__input_summary.details, details)
self.assertEqual(info._StateInfo__output_summary.brief, "-")
self.assertEqual(info._StateInfo__output_summary.details, no_output)
# send template data
self.send_signal(self.widget.Inputs.template_data, self.disc_data)
output = self.get_output(self.widget.Outputs.transformed_data)
self.assertTableEqual(output, self.disc_data[::15])
self.assertEqual("Input data with 10 instances and 4 features.",
self.widget.input_label.text())
self.assertEqual("Template domain applied.",
self.widget.template_label.text())
self.assertEqual("Output data includes 4 features.",
self.widget.output_label.text())
# remove data
self.send_signal(self.widget.Inputs.data, None)
output = self.get_output(self.widget.Outputs.transformed_data)
self.assertIsNone(output)
self.assertEqual("No data on input.", self.widget.input_label.text())
self.assertEqual("Template data includes 4 features.",
self.widget.template_label.text())
self.assertEqual("", self.widget.output_label.text())
data_list = [("Data", None), ("Template data", self.disc_data)]
summary, details = "0, 150", format_multiple_summaries(data_list)
self.assertEqual(info._StateInfo__input_summary.brief, summary)
self.assertEqual(info._StateInfo__input_summary.details, details)
self.assertEqual(info._StateInfo__output_summary.brief, "-")
self.assertEqual(info._StateInfo__output_summary.details, no_output)
# remove template data
self.send_signal(self.widget.Inputs.template_data, None)
self.assertEqual("No data on input.", self.widget.input_label.text())
self.assertEqual("No template data on input.",
self.widget.template_label.text())
self.assertEqual("", self.widget.output_label.text())
self.assertEqual(info._StateInfo__input_summary.brief, "-")
self.assertEqual(info._StateInfo__input_summary.details, no_input)
self.assertEqual(info._StateInfo__output_summary.brief, "-")
self.assertEqual(info._StateInfo__output_summary.details, no_output)
def test_summary(self):
"""Check if status bar is updated when data is received"""
info = self.widget.info
no_input, no_output = "No data on input", "No data on output"
zoo = Table("zoo")
data_list = [("zoo", zoo)]
self.send_signal(self.widget.Inputs.data, zoo, 1)
summary, details = "101", format_multiple_summaries(data_list)
self.assertEqual(info._StateInfo__input_summary.brief, summary)
self.assertEqual(info._StateInfo__input_summary.details, details)
self.assertEqual(info._StateInfo__output_summary.brief, "")
self.assertEqual(info._StateInfo__output_summary.details, no_output)
self._select_data()
output = self.get_output(self.widget.Outputs.selected_data)
summary, details = f"{len(output)}", format_summary_details(output)
self.assertEqual(info._StateInfo__output_summary.brief, summary)
self.assertEqual(info._StateInfo__output_summary.details, details)
iris = Table("iris")
data_list = [("zoo", zoo), ("iris", iris)]
self.send_signal(self.widget.Inputs.data, iris, 2)
summary, details = "101, 150", format_multiple_summaries(data_list)
self.assertEqual(info._StateInfo__input_summary.brief, summary)
self.assertEqual(info._StateInfo__input_summary.details, details)
self.assertEqual(info._StateInfo__output_summary.brief, "")
self.assertEqual(info._StateInfo__output_summary.details, no_output)
self._select_data()
output = self.get_output(self.widget.Outputs.selected_data)
summary, details = f"{len(output)}", format_summary_details(output)
self.assertEqual(info._StateInfo__output_summary.brief, summary)
self.assertEqual(info._StateInfo__output_summary.details, details)
brown = Table("brown-selected")
data_list = [("zoo", zoo), ("iris", iris), ("brown-selected", brown)]
self.send_signal(self.widget.Inputs.data, brown, 3)
summary, details = "101, 150, 186", format_multiple_summaries(data_list)
self.assertEqual(info._StateInfo__input_summary.brief, summary)
self.assertEqual(info._StateInfo__input_summary.details, details)
self._select_data()
output = self.get_output(self.widget.Outputs.selected_data)
summary, details = f"{len(output)}", format_summary_details(output)
self.assertEqual(info._StateInfo__output_summary.brief, summary)
self.assertEqual(info._StateInfo__output_summary.details, details)
self.send_signal(self.widget.Inputs.data, None, 1)
data_list = [("iris", iris), ("brown-selected", brown)]
summary, details = "150, 186", format_multiple_summaries(data_list)
self.assertEqual(info._StateInfo__input_summary.brief, summary)
self.assertEqual(info._StateInfo__input_summary.details, details)
self.assertEqual(info._StateInfo__output_summary.brief, "")
self.assertEqual(info._StateInfo__output_summary.details, no_output)
self._select_data()
output = self.get_output(self.widget.Outputs.selected_data)
summary, details = f"{len(output)}", format_summary_details(output)
self.assertEqual(info._StateInfo__output_summary.brief, summary)
self.assertEqual(info._StateInfo__output_summary.details, details)
self.send_signal(self.widget.Inputs.data, None, 2)
self.send_signal(self.widget.Inputs.data, None, 3)
self.assertEqual(info._StateInfo__input_summary.brief, "")
self.assertEqual(info._StateInfo__input_summary.details, no_input)
self.assertEqual(info._StateInfo__output_summary.brief, "")
self.assertEqual(info._StateInfo__output_summary.details, no_output)
def __commit_finish(self):
assert QThread.currentThread() is self.thread()
assert self.__task is not None
futures = self.__task.futures
assert len(futures) == len(self.varmodel)
assert self.data is not None
def get_variable(variable, future, drop_mask) \
-> Optional[List[Orange.data.Variable]]:
# Returns a (potentially empty) list of variables,
# or None on failure that should interrupt the imputation
assert future.done()
try:
res = future.result()
except SparseNotSupported:
self.Error.model_based_imputer_sparse()
return [] # None?
except VariableNotSupported:
self.Warning.cant_handle_var(variable.name)
return []
except Exception: # pylint: disable=broad-except
log = logging.getLogger(__name__)
log.info("Error for %s", variable.name, exc_info=True)
self.Error.imputation_failed(variable.name)
return None
if isinstance(res, RowMask):
drop_mask |= res.mask
newvar = variable
else:
newvar = res
if isinstance(newvar, Orange.data.Variable):
newvar = [newvar]
return newvar
def create_data(attributes, class_vars):
domain = Orange.data.Domain(attributes, class_vars,
self.data.domain.metas)
try:
return self.data.from_table(domain, self.data[~drop_mask])
except Exception: # pylint: disable=broad-except
log = logging.getLogger(__name__)
log.info("Error", exc_info=True)
self.Error.imputation_failed("Unknown")
return None
self.__task = None
self.setInvalidated(False)
self.progressBarFinished()
attributes = []
class_vars = []
drop_mask = np.zeros(len(self.data), bool)
for i, (var, fut) in enumerate(zip(self.varmodel, futures)):
newvar = get_variable(var, fut, drop_mask)
if newvar is None:
data = None
break
if i < len(self.data.domain.attributes):
attributes.extend(newvar)
else:
class_vars.extend(newvar)
else:
data = create_data(attributes, class_vars)
self.Outputs.data.send(data)
self.modified = False
summary = len(data) if data else self.info.NoOutput
details = format_summary_details(data) if data else ""
self.info.set_output_summary(summary, details)
def test_details(self):
"""Check if details part of the summary is formatted correctly"""
data = Table('zoo')
n_features = len(data.domain.variables) + len(data.domain.metas)
details = f'{len(data)} instances, ' \
f'{n_features} features\n' \
f'Features: {len(data.domain.attributes)} categorical\n' \
f'Target: categorical\n' \
f'Metas: string (not shown)'
self.assertEqual(details, format_summary_details(data))
data = Table('housing')
n_features = len(data.domain.variables) + len(data.domain.metas)
details = f'{len(data)} instances, ' \
f'{n_features} features\n' \
f'Features: {len(data.domain.attributes)} numeric\n' \
f'Target: numeric\n' \
f'Metas: —'
self.assertEqual(details, format_summary_details(data))
data = Table('heart_disease')
n_features = len(data.domain.variables) + len(data.domain.metas)
details = f'{len(data)} instances, ' \
f'{n_features} features\n' \
f'Features: {len(data.domain.attributes)} ' \
f'(7 categorical, 6 numeric)\n' \
f'Target: categorical\n' \
f'Metas: —'
self.assertEqual(details, format_summary_details(data))
data = make_table([continuous_full, continuous_missing],
target=[rgb_full, rgb_missing],
metas=[ints_full, ints_missing])
n_features = len(data.domain.variables) + len(data.domain.metas)
details = f'{len(data)} instances, ' \
f'{n_features} features\n' \
f'Features: {len(data.domain.attributes)} numeric\n' \
f'Target: {len(data.domain.class_vars)} categorical\n' \
f'Metas: {len(data.domain.metas)} categorical'
self.assertEqual(details, format_summary_details(data))
data = make_table([continuous_full, time_full, ints_full, rgb_missing],
target=[rgb_full, continuous_missing],
metas=[string_full, string_missing])
n_features = len(data.domain.variables) + len(data.domain.metas)
details = f'{len(data)} instances, ' \
f'{n_features} features\n' \
f'Features: {len(data.domain.attributes)} ' \
f'(2 categorical, 1 numeric, 1 time)\n' \
f'Target: {len(data.domain.class_vars)} ' \
f'(1 categorical, 1 numeric)\n' \
f'Metas: {len(data.domain.metas)} string (not shown)'
self.assertEqual(details, format_summary_details(data))
data = make_table([time_full, time_missing],
target=[ints_missing],
metas=None)
details = f'{len(data)} instances, ' \
f'{len(data.domain.variables)} features\n' \
f'Features: {len(data.domain.attributes)} time\n'\
f'Target: categorical\n' \
f'Metas: —'
self.assertEqual(details, format_summary_details(data))
data = make_table([rgb_full, ints_full], target=None, metas=None)
details = f'{len(data)} instances, ' \
f'{len(data.domain.variables)} features\n' \
f'Features: {len(data.domain.variables)} categorical\n' \
f'Target: —\n' \
f'Metas: —'
self.assertEqual(details, format_summary_details(data))
data = make_table([rgb_full], target=None, metas=None)
details = f'{len(data)} instances, ' \
f'{len(data.domain.variables)} feature\n' \
f'Features: categorical\n' \
f'Target: —\n' \
f'Metas: —'
self.assertEqual(details, format_summary_details(data))
data = None
self.assertEqual('', format_summary_details(data))
def set_data(self, data: Optional[Table]):
self.data = data
summary = len(data) if data else self.info.NoInput
details = format_summary_details(data) if data else ""
self.info.set_input_summary(summary, details)
def _set_input_summary(self, dataset):
summary = len(dataset) if dataset else self.info.NoInput
details = format_summary_details(dataset) if dataset else ""
self.info.set_input_summary(summary, details)
def _set_summary(data, empty, setter):
summary = len(data) if data else empty
details = format_summary_details(data) if data else ""
setter(summary, details)
def set_data(self, data=None):
self.update_domain_role_hints()
self.closeContext()
self.data = data
if data is not None:
self.openContext(data)
all_vars = data.domain.variables + data.domain.metas
var_sig = lambda attr: (attr.name, vartype(attr))
domain_hints = {
var_sig(attr): ("attribute", i)
for i, attr in enumerate(data.domain.attributes)
}
domain_hints.update({
var_sig(attr): ("meta", i)
for i, attr in enumerate(data.domain.metas)
})
if data.domain.class_vars:
domain_hints.update({
var_sig(attr): ("class", i)
for i, attr in enumerate(data.domain.class_vars)
})
# update the hints from context settings
domain_hints.update(self.domain_role_hints)
attrs_for_role = lambda role: [
(domain_hints[var_sig(attr)][1], attr) for attr in all_vars
if domain_hints[var_sig(attr)][0] == role
]
attributes = [
attr for place, attr in sorted(attrs_for_role("attribute"),
key=lambda a: a[0])
]
classes = [
attr for place, attr in sorted(attrs_for_role("class"),
key=lambda a: a[0])
]
metas = [
attr for place, attr in sorted(attrs_for_role("meta"),
key=lambda a: a[0])
]
available = [
attr for place, attr in sorted(attrs_for_role("available"),
key=lambda a: a[0])
]
self.used_attrs[:] = attributes
self.class_attrs[:] = classes
self.meta_attrs[:] = metas
self.available_attrs[:] = available
self.info.set_input_summary(len(data),
format_summary_details(data))
else:
self.used_attrs[:] = []
self.class_attrs[:] = []
self.meta_attrs[:] = []
self.available_attrs[:] = []
self.info.set_input_summary(self.info.NoInput)
def set_data(self, data):
self.data = data
summary = len(data) if data else self.info.NoInput
details = format_summary_details(data) if data else ""
self.info.set_input_summary(summary, details)
self.unconditional_apply()
