class OWLearningCurveB(OWBaseWidget):
name = "Learning Curve (B)"
description = ("Takes a dataset and a set of learners and shows a "
"learning curve in a table")
icon = "icons/LearningCurve.svg"
priority = 1010
# [start-snippet-1]
class Inputs:
data = Input("Data", Table, default=True)
test_data = Input("Test Data", Table)
learner = MultiInput("Learner", Learner)
# [end-snippet-1]
#: cross validation folds
folds = settings.Setting(5)
#: points in the learning curve
steps = settings.Setting(10)
#: index of the selected scoring function
scoringF = settings.Setting(0)
#: compute curve on any change of parameters
commitOnChange = settings.Setting(True)
def __init__(self):
super().__init__()
# sets self.curvePoints, self.steps equidistant points from
# 1/self.steps to 1
self.updateCurvePoints()
self.scoring = [("Classification Accuracy",
Orange.evaluation.scoring.CA),
("AUC", Orange.evaluation.scoring.AUC),
("Precision", Orange.evaluation.scoring.Precision),
("Recall", Orange.evaluation.scoring.Recall)]
#: Input data on which to construct the learning curve
self.data = None
#: Optional test data
self.testdata = None
#: LearnerData for each learner input
self.learners: List[LearnerData] = []
# GUI
box = gui.widgetBox(self.controlArea, "Info")
self.infoa = gui.widgetLabel(box, 'No data on input.')
self.infob = gui.widgetLabel(box, 'No learners.')
gui.separator(self.controlArea)
box = gui.widgetBox(self.controlArea, "Evaluation Scores")
gui.comboBox(box,
self,
"scoringF",
items=[x[0] for x in self.scoring],
callback=self._invalidate_curves)
gui.separator(self.controlArea)
box = gui.widgetBox(self.controlArea, "Options")
gui.spin(box,
self,
'folds',
2,
100,
step=1,
label='Cross validation folds: ',
keyboardTracking=False,
callback=lambda: self._invalidate_results()
if self.commitOnChange else None)
gui.spin(box,
self,
'steps',
2,
100,
step=1,
label='Learning curve points: ',
keyboardTracking=False,
callback=[
self.updateCurvePoints,
lambda: self._invalidate_results()
if self.commitOnChange else None
])
gui.checkBox(box, self, 'commitOnChange',
'Apply setting on any change')
self.commitBtn = gui.button(box,
self,
"Apply Setting",
callback=self._invalidate_results,
disabled=True)
gui.rubber(self.controlArea)
# table widget
self.table = gui.table(self.mainArea,
selectionMode=QTableWidget.NoSelection)
##########################################################################
# slots: handle input signals
@Inputs.data
def set_dataset(self, data):
"""Set the input train dataset."""
# Clear all results/scores for all learner inputs
for item in self.learners:
item.results = None
item.curve = None
self.data = data
if data is not None:
self.infoa.setText('%d instances in input dataset' % len(data))
else:
self.infoa.setText('No data on input.')
self.commitBtn.setEnabled(self.data is not None)
@Inputs.test_data
def set_testdataset(self, testdata):
"""Set a separate test dataset."""
# Clear all results/scores for all learner inputs
for item in self.learners:
item.results = None
item.curve = None
self.testdata = testdata
@Inputs.learner
def set_learner(self, index: int, learner):
"""Set the input learner at index"""
# update/replace a learner on a previously connected link
item = self.learners[index]
item.learner = learner
item.results = None
item.curve = None
@Inputs.learner.insert
def insert_learner(self, index, learner):
"""Insert a learner at index"""
self.learners.insert(index, LearnerData(learner, None, None))
@Inputs.learner.remove
def remove_learner(self, index):
""""Remove a learner at index"""
# remove a learner and corresponding results
del self.learners[index]
def handleNewSignals(self):
if len(self.learners):
self.infob.setText("%d learners on input." % len(self.learners))
else:
self.infob.setText("No learners.")
self.commitBtn.setEnabled(len(self.learners))
if self.data is not None:
self._update()
self._update_curve_points()
self._update_table()
def _invalidate_curves(self):
if self.data is not None:
self._update_curve_points()
self._update_table()
def _invalidate_results(self):
for item in self.learners:
item.results = None
item.curve = None
if self.data is not None:
self._update()
self._update_curve_points()
self._update_table()
def _update(self):
assert self.data is not None
# collect all learners for which results have not yet been computed
need_update = [(i, item) for (i, item) in enumerate(self.learners)
if item.results is None]
if not need_update:
return
learners = [item.learner for _, item in need_update]
if self.testdata is None:
# compute the learning curve result for all learners in one go
results = learning_curve(
learners,
self.data,
folds=self.folds,
proportions=self.curvePoints,
)
else:
results = learning_curve_with_test_data(
learners,
self.data,
self.testdata,
times=self.folds,
proportions=self.curvePoints,
)
# split the combined result into per learner/model results
results = [
list(Results.split_by_model(p_results)) for p_results in results
]
for i, (_, item) in enumerate(need_update):
item.results = [p_results[i] for p_results in results]
def _update_curve_points(self):
scoref = self.scoring[self.scoringF][1]
for item in self.learners:
item.curve = [scoref(x)[0] for x in item.results]
def _update_table(self):
self.table.setRowCount(0)
self.table.setRowCount(len(self.curvePoints))
self.table.setColumnCount(len(self.learners))
self.table.setHorizontalHeaderLabels(
[item.learner.name for item in self.learners])
self.table.setVerticalHeaderLabels(
["{:.2f}".format(p) for p in self.curvePoints])
if self.data is None:
return
for column, item in enumerate(self.learners):
for row, point in enumerate(item.curve):
self.table.setItem(row, column,
QTableWidgetItem("{:.5f}".format(point)))
for i in range(len(self.learners)):
sh = self.table.sizeHintForColumn(i)
cwidth = self.table.columnWidth(i)
self.table.setColumnWidth(i, max(sh, cwidth))
def updateCurvePoints(self):
self.curvePoints = [(x + 1.) / self.steps for x in range(self.steps)]
class OWLearningCurveC(OWBaseWidget):
name = "Learning Curve (C)"
description = ("Takes a dataset and a set of learners and shows a "
"learning curve in a table")
icon = "icons/LearningCurve.svg"
priority = 1010
class Inputs:
data = Input("Data", Table, default=True)
test_data = Input("Test Data", Table)
learner = MultiInput("Learner", Learner)
#: cross validation folds
folds = settings.Setting(5)
#: points in the learning curve
steps = settings.Setting(10)
#: index of the selected scoring function
scoringF = settings.Setting(0)
#: compute curve on any change of parameters
commitOnChange = settings.Setting(True)
def __init__(self):
super().__init__()
# sets self.curvePoints, self.steps equidistant points from
# 1/self.steps to 1
self.updateCurvePoints()
self.scoring = [("Classification Accuracy",
Orange.evaluation.scoring.CA),
("AUC", Orange.evaluation.scoring.AUC),
("Precision", Orange.evaluation.scoring.Precision),
("Recall", Orange.evaluation.scoring.Recall)]
#: Input data on which to construct the learning curve
self.data = None
#: Optional test data
self.testdata = None
#: LearnerData for each learner input
self.learners: List[LearnerData] = []
# [start-snippet-3]
#: The current evaluating task (if any)
self._task = None # type: Optional[Task]
#: An executor we use to submit learner evaluations into a thread pool
self._executor = concurrent.futures.ThreadPoolExecutor()
# [end-snippet-3]
# GUI
box = gui.widgetBox(self.controlArea, "Info")
self.infoa = gui.widgetLabel(box, 'No data on input.')
self.infob = gui.widgetLabel(box, 'No learners.')
gui.separator(self.controlArea)
box = gui.widgetBox(self.controlArea, "Evaluation Scores")
gui.comboBox(box,
self,
"scoringF",
items=[x[0] for x in self.scoring],
callback=self._invalidate_curves)
gui.separator(self.controlArea)
box = gui.widgetBox(self.controlArea, "Options")
gui.spin(box,
self,
'folds',
2,
100,
step=1,
label='Cross validation folds: ',
keyboardTracking=False,
callback=lambda: self._invalidate_results()
if self.commitOnChange else None)
gui.spin(box,
self,
'steps',
2,
100,
step=1,
label='Learning curve points: ',
keyboardTracking=False,
callback=[
self.updateCurvePoints,
lambda: self._invalidate_results()
if self.commitOnChange else None
])
gui.checkBox(box, self, 'commitOnChange',
'Apply setting on any change')
self.commitBtn = gui.button(box,
self,
"Apply Setting",
callback=self._invalidate_results,
disabled=True)
gui.rubber(self.controlArea)
# table widget
self.table = gui.table(self.mainArea,
selectionMode=QTableWidget.NoSelection)
##########################################################################
# slots: handle input signals
@Inputs.data
def set_dataset(self, data):
"""Set the input train dataset."""
# Clear all results/scores
for item in self.learners:
item.results = None
item.curve = None
self.data = data
if data is not None:
self.infoa.setText('%d instances in input dataset' % len(data))
else:
self.infoa.setText('No data on input.')
self.commitBtn.setEnabled(self.data is not None)
@Inputs.test_data
def set_testdataset(self, testdata):
"""Set a separate test dataset."""
# Clear all results/scores for all learner inputs
for item in self.learners:
item.results = None
item.curve = None
self.testdata = testdata
@Inputs.learner
def set_learner(self, index: int, learner):
"""Set the input learner at index"""
# update/replace a learner on a previously connected link
item = self.learners[index]
item.learner = learner
item.results = None
item.curve = None
@Inputs.learner.insert
def insert_learner(self, index, learner):
"""Insert a learner at index"""
self.learners.insert(index, LearnerData(learner, None, None))
@Inputs.learner.remove
def remove_learner(self, index):
""""Remove a learner at index"""
# remove a learner and corresponding results
del self.learners[index]
# [start-snippet-4]
def handleNewSignals(self):
if len(self.learners):
self.infob.setText("%d learners on input." % len(self.learners))
else:
self.infob.setText("No learners.")
self.commitBtn.setEnabled(len(self.learners))
self._update()
# [end-snippet-4]
def _invalidate_curves(self):
if self.data is not None:
self._update_curve_points()
self._update_table()
def _invalidate_results(self):
for item in self.learners:
item.results = None
item.curve = None
self._update()
# [start-snippet-5]
def _update(self):
if self._task is not None:
# First make sure any pending tasks are cancelled.
self.cancel()
assert self._task is None
if self.data is None:
return
# collect all learners for which results have not yet been computed
need_update = [(i, item) for (i, item) in enumerate(self.learners)
if item.results is None]
if not need_update:
self._update_curve_points()
self._update_table()
return
# [end-snippet-5]
# [start-snippet-6]
learners = [item.learner for _, item in need_update]
# setup the learner evaluations as partial function capturing
# the necessary arguments.
if self.testdata is None:
learning_curve_func = partial(
learning_curve,
learners,
self.data,
folds=self.folds,
proportions=self.curvePoints,
)
else:
learning_curve_func = partial(
learning_curve_with_test_data,
learners,
self.data,
self.testdata,
times=self.folds,
proportions=self.curvePoints,
)
# [end-snippet-6]
# [start-snippet-7]
# setup the task state
self._task = task = Task()
# The learning_curve[_with_test_data] also takes a callback function
# to report the progress. We instrument this callback to both invoke
# the appropriate slots on this widget for reporting the progress
# (in a thread safe manner) and to implement cooperative cancellation.
set_progress = methodinvoke(self, "setProgressValue", (float, ))
def callback(finished):
# check if the task has been cancelled and raise an exception
# from within. This 'strategy' can only be used with code that
# properly cleans up after itself in the case of an exception
# (does not leave any global locks, opened file descriptors, ...)
if task.cancelled:
raise KeyboardInterrupt()
set_progress(finished * 100)
# capture the callback in the partial function
learning_curve_func = partial(learning_curve_func, callback=callback)
# [end-snippet-7]
# [start-snippet-8]
self.progressBarInit()
# Submit the evaluation function to the executor and fill in the
# task with the resultant Future.
task.future = self._executor.submit(learning_curve_func)
# Setup the FutureWatcher to notify us of completion
task.watcher = FutureWatcher(task.future)
# by using FutureWatcher we ensure `_task_finished` slot will be
# called from the main GUI thread by the Qt's event loop
task.watcher.done.connect(self._task_finished)
# [end-snippet-8]
# [start-snippet-progress]
@pyqtSlot(float)
def setProgressValue(self, value):
assert self.thread() is QThread.currentThread()
self.progressBarSet(value)
# [end-snippet-progress]
# [start-snippet-9]
@pyqtSlot(concurrent.futures.Future)
def _task_finished(self, f):
"""
Parameters
----------
f : Future
The future instance holding the result of learner evaluation.
"""
assert self.thread() is QThread.currentThread()
assert self._task is not None
assert self._task.future is f
assert f.done()
self._task = None
self.progressBarFinished()
try:
results = f.result() # type: List[Results]
except Exception as ex:
# Log the exception with a traceback
log = logging.getLogger()
log.exception(__name__, exc_info=True)
self.error("Exception occurred during evaluation: {!r}".format(ex))
# clear all results
for item in self.learners:
item.results = None
else:
# split the combined result into per learner/model results ...
results = [
list(Results.split_by_model(p_results))
for p_results in results
] # type: List[List[Results]]
assert all(len(r.learners) == 1 for r1 in results for r in r1)
assert len(results) == len(self.curvePoints)
learners = [r.learners[0] for r in results[0]]
# map learner back to LearnerData instance
data_by_learner = {item.learner: item for item in self.learners}
# ... and update self.results
for i, learner in enumerate(learners):
item = data_by_learner[learner]
item.results = [p_results[i] for p_results in results]
# update the display
self._update_curve_points()
self._update_table()
# [end-snippet-9]
# [start-snippet-10]
def cancel(self):
"""
Cancel the current task (if any).
"""
if self._task is not None:
self._task.cancel()
assert self._task.future.done()
# disconnect the `_task_finished` slot
self._task.watcher.done.disconnect(self._task_finished)
self._task = None
self.progressBarFinished()
# [end-snippet-10]
# [start-snippet-11]
def onDeleteWidget(self):
self.cancel()
super().onDeleteWidget()
# [end-snippet-11]
def _update_curve_points(self):
scoref = self.scoring[self.scoringF][1]
for item in self.learners:
item.curve = [scoref(x)[0] for x in item.results]
def _update_table(self):
self.table.setRowCount(0)
self.table.setRowCount(len(self.curvePoints))
self.table.setColumnCount(len(self.learners))
self.table.setHorizontalHeaderLabels(
[item.learner.name for item in self.learners])
self.table.setVerticalHeaderLabels(
["{:.2f}".format(p) for p in self.curvePoints])
if self.data is None:
return
for column, item in enumerate(self.learners):
for row, point in enumerate(item.curve):
self.table.setItem(row, column,
QTableWidgetItem("{:.5f}".format(point)))
for i in range(len(self.learners)):
sh = self.table.sizeHintForColumn(i)
cwidth = self.table.columnWidth(i)
self.table.setColumnWidth(i, max(sh, cwidth))
def updateCurvePoints(self):
self.curvePoints = [(x + 1.) / self.steps for x in range(self.steps)]
class OWNxEmbedding(OWWidget):
name = "Network Embeddings"
description = "Embed network elements"
icon = "icons/NetworkEmbedding.svg"
priority = 6450
class Inputs:
network = Input("Network", Network, default=True)
class Outputs:
items = Output("Items", Table)
resizing_enabled = False
want_main_area = False
p = settings.Setting(1.0)
q = settings.Setting(1.0)
walk_len = settings.Setting(50)
num_walks = settings.Setting(10)
emb_size = settings.Setting(128)
window_size = settings.Setting(5)
num_epochs = settings.Setting(1)
auto_commit = settings.Setting(True)
def __init__(self):
super().__init__()
self.network = None
self.embedder = None
self._worker_thread = None
self._progress_updater = None
def commit():
return self.commit()
box = gui.widgetBox(self.controlArea, box=True)
kwargs = dict(controlWidth=75,
alignment=Qt.AlignRight,
callback=commit)
gui.spin(box,
self,
"p",
0.0,
10.0,
0.1,
label="Return parameter (p): ",
spinType=float,
**kwargs)
gui.spin(box,
self,
"q",
0.0,
10.0,
0.1,
label="In-out parameter (q): ",
spinType=float,
**kwargs)
gui.spin(box,
self,
"walk_len",
1,
100_000,
1,
label="Walk length: ",
**kwargs)
gui.spin(box,
self,
"num_walks",
1,
10_000,
1,
label="Walks per node: ",
**kwargs)
gui.spin(box,
self,
"emb_size",
1,
10_000,
1,
label="Embedding size: ",
**kwargs)
gui.spin(box,
self,
"window_size",
1,
20,
1,
label="Context size: ",
**kwargs)
gui.spin(box,
self,
"num_epochs",
1,
100,
1,
label="Number of epochs: ",
**kwargs)
gui.auto_commit(self.controlArea,
self,
"auto_commit",
"Commit",
checkbox_label="Auto-commit",
orientation=Qt.Horizontal)
commit()
@Inputs.network
def set_network(self, net):
self.network = net
self.commit()
def commit(self):
self.Warning.clear()
# cancel existing computation if running
if self._worker_thread is not None:
self._worker_thread.finished.disconnect()
self._worker_thread.quit()
self._worker_thread = None
if self.network is None:
self.Outputs.items.send(None)
return
self._progress_updater = ProgressBarUpdater(self, self.num_epochs)
self.embedder = embeddings.Node2Vec(self.p,
self.q,
self.walk_len,
self.num_walks,
self.emb_size,
self.window_size,
self.num_epochs,
callbacks=[self._progress_updater])
self._worker_thread = EmbedderThread(
lambda: self.embedder(self.network))
self._worker_thread.finished.connect(self.on_finished)
self.progressBarInit()
self.progressBarSet(1e-5)
self._worker_thread.start()
def on_finished(self):
output = self._worker_thread.result
self._worker_thread = None
self._progress_updater = None
self.progressBarFinished()
self.Outputs.items.send(output)
def onDeleteWidget(self):
if self._worker_thread is not None:
# prevent the callback from trying to access deleted widget object
self._progress_updater.widget = None
self._worker_thread.finished.disconnect()
self._worker_thread.quit()
super().onDeleteWidget()
class OWLearningCurveB(OWBaseWidget):
name = "Learning Curve (B)"
description = ("Takes a dataset and a set of learners and shows a "
"learning curve in a table")
icon = "icons/LearningCurve.svg"
priority = 1010
# [start-snippet-1]
class Inputs:
data = Input("Data", Orange.data.Table, default=True)
test_data = Input("Test Data", Orange.data.Table)
learner = Input("Learner",
Orange.classification.Learner,
multiple=True)
# [end-snippet-1]
#: cross validation folds
folds = settings.Setting(5)
#: points in the learning curve
steps = settings.Setting(10)
#: index of the selected scoring function
scoringF = settings.Setting(0)
#: compute curve on any change of parameters
commitOnChange = settings.Setting(True)
def __init__(self):
super().__init__()
# sets self.curvePoints, self.steps equidistant points from
# 1/self.steps to 1
self.updateCurvePoints()
self.scoring = [("Classification Accuracy",
Orange.evaluation.scoring.CA),
("AUC", Orange.evaluation.scoring.AUC),
("Precision", Orange.evaluation.scoring.Precision),
("Recall", Orange.evaluation.scoring.Recall)]
#: input data on which to construct the learning curve
self.data = None
#: optional test data
self.testdata = None
#: A {input_id: Learner} mapping of current learners from input channel
self.learners = OrderedDict()
#: A {input_id: List[Results]} mapping of input id to evaluation
#: results list, one for each curve point
self.results = OrderedDict()
#: A {input_id: List[float]} mapping of input id to learning curve
#: point scores
self.curves = OrderedDict()
# GUI
box = gui.widgetBox(self.controlArea, "Info")
self.infoa = gui.widgetLabel(box, 'No data on input.')
self.infob = gui.widgetLabel(box, 'No learners.')
gui.separator(self.controlArea)
box = gui.widgetBox(self.controlArea, "Evaluation Scores")
gui.comboBox(box,
self,
"scoringF",
items=[x[0] for x in self.scoring],
callback=self._invalidate_curves)
gui.separator(self.controlArea)
box = gui.widgetBox(self.controlArea, "Options")
gui.spin(box,
self,
'folds',
2,
100,
step=1,
label='Cross validation folds: ',
keyboardTracking=False,
callback=lambda: self._invalidate_results()
if self.commitOnChange else None)
gui.spin(box,
self,
'steps',
2,
100,
step=1,
label='Learning curve points: ',
keyboardTracking=False,
callback=[
self.updateCurvePoints,
lambda: self._invalidate_results()
if self.commitOnChange else None
])
gui.checkBox(box, self, 'commitOnChange',
'Apply setting on any change')
self.commitBtn = gui.button(box,
self,
"Apply Setting",
callback=self._invalidate_results,
disabled=True)
gui.rubber(self.controlArea)
# table widget
self.table = gui.table(self.mainArea,
selectionMode=QTableWidget.NoSelection)
##########################################################################
# slots: handle input signals
@Inputs.data
def set_dataset(self, data):
"""Set the input train dataset."""
# Clear all results/scores
for id in list(self.results):
self.results[id] = None
for id in list(self.curves):
self.curves[id] = None
self.data = data
if data is not None:
self.infoa.setText('%d instances in input dataset' % len(data))
else:
self.infoa.setText('No data on input.')
self.commitBtn.setEnabled(self.data is not None)
@Inputs.test_data
def set_testdataset(self, testdata):
"""Set a separate test dataset."""
# Clear all results/scores
for id in list(self.results):
self.results[id] = None
for id in list(self.curves):
self.curves[id] = None
self.testdata = testdata
@Inputs.learner
def set_learner(self, learner, id):
"""Set the input learner for channel id."""
if id in self.learners:
if learner is None:
# remove a learner and corresponding results
del self.learners[id]
del self.results[id]
del self.curves[id]
else:
# update/replace a learner on a previously connected link
self.learners[id] = learner
# invalidate the cross-validation results and curve scores
# (will be computed/updated in `_update`)
self.results[id] = None
self.curves[id] = None
else:
if learner is not None:
self.learners[id] = learner
# initialize the cross-validation results and curve scores
# (will be computed/updated in `_update`)
self.results[id] = None
self.curves[id] = None
if len(self.learners):
self.infob.setText("%d learners on input." % len(self.learners))
else:
self.infob.setText("No learners.")
self.commitBtn.setEnabled(len(self.learners))
def handleNewSignals(self):
if self.data is not None:
self._update()
self._update_curve_points()
self._update_table()
def _invalidate_curves(self):
if self.data is not None:
self._update_curve_points()
self._update_table()
def _invalidate_results(self):
for id in self.learners:
self.curves[id] = None
self.results[id] = None
if self.data is not None:
self._update()
self._update_curve_points()
self._update_table()
def _update(self):
assert self.data is not None
# collect all learners for which results have not yet been computed
need_update = [(id, learner) for id, learner in self.learners.items()
if self.results[id] is None]
if not need_update:
return
learners = [learner for _, learner in need_update]
if self.testdata is None:
# compute the learning curve result for all learners in one go
results = learning_curve(
learners,
self.data,
folds=self.folds,
proportions=self.curvePoints,
)
else:
results = learning_curve_with_test_data(
learners,
self.data,
self.testdata,
times=self.folds,
proportions=self.curvePoints,
)
# split the combined result into per learner/model results
results = [
list(Results.split_by_model(p_results)) for p_results in results
]
for i, (id, learner) in enumerate(need_update):
self.results[id] = [p_results[i] for p_results in results]
def _update_curve_points(self):
for id in self.learners:
curve = [
self.scoring[self.scoringF][1](x)[0] for x in self.results[id]
]
self.curves[id] = curve
def _update_table(self):
self.table.setRowCount(0)
self.table.setRowCount(len(self.curvePoints))
self.table.setColumnCount(len(self.learners))
self.table.setHorizontalHeaderLabels(
[learner.name for _, learner in self.learners.items()])
self.table.setVerticalHeaderLabels(
["{:.2f}".format(p) for p in self.curvePoints])
if self.data is None:
return
for column, curve in enumerate(self.curves.values()):
for row, point in enumerate(curve):
self.table.setItem(row, column,
QTableWidgetItem("{:.5f}".format(point)))
for i in range(len(self.learners)):
sh = self.table.sizeHintForColumn(i)
cwidth = self.table.columnWidth(i)
self.table.setColumnWidth(i, max(sh, cwidth))
def updateCurvePoints(self):
self.curvePoints = [(x + 1.) / self.steps for x in range(self.steps)]
# [start-snippet-3]
def test_run_signals(self):
data = Orange.data.Table("iris")
indices = numpy.random.permutation(len(data))
traindata = data[indices[:-20]]
testdata = data[indices[-20:]]
self.set_dataset(traindata)
self.set_testdataset(testdata)
l1 = Orange.classification.NaiveBayesLearner()
l1.name = 'Naive Bayes'
self.set_learner(l1, 1)
l2 = Orange.classification.LogisticRegressionLearner()
l2.name = 'Logistic Regression'
self.set_learner(l2, 2)
l4 = Orange.classification.SklTreeLearner()
l4.name = "Decision Tree"
self.set_learner(l4, 3)
