def randomize(self, data_matrix, amount=.5):
random.seed(self.random_state)
inclusion_threshold = random.uniform(amount, 1)
selectors = []
if random.random() > inclusion_threshold:
selectors.append(
QuickShiftSelector(random_state=random.randint(1, 1e9)))
if random.random() > inclusion_threshold:
selectors.append(
DecisionSurfaceSelector(random_state=random.randint(1, 1e9)))
if random.random() > inclusion_threshold:
selectors.append(
SparseSelector(random_state=random.randint(1, 1e9)))
if random.random() > inclusion_threshold:
selectors.append(
MaxVolSelector(random_state=random.randint(1, 1e9)))
if random.random() > inclusion_threshold:
selectors.append(
DensitySelector(random_state=random.randint(1, 1e9)))
if random.random() > inclusion_threshold:
selectors.append(
OnionSelector(random_state=random.randint(1, 1e9)))
if not selectors:
selectors.append(
QuickShiftSelector(random_state=random.randint(1, 1e9)))
self.selector = CompositeSelector(selectors=selectors)
self.selector.randomize(data_matrix, amount=amount)
self.metric = 'rbf'
self.kwds = {'gamma': random.choice([10**x for x in range(-3, 3)])}
self.random_state = self.random_state ^ random.randint(1, 1e9)
def make_selectors(opts):
# compose selector list
selectors = []
if 'QuickShiftSelector' in opts and opts['QuickShiftSelector'] is not None:
from eden.selector import QuickShiftSelector
selectors.append(
QuickShiftSelector(n_instances=opts['QuickShiftSelector']))
if 'MaxVolSelector' in opts and opts['MaxVolSelector'] is not None:
from eden.selector import MaxVolSelector
selectors.append(MaxVolSelector(n_instances=opts['MaxVolSelector']))
if 'DensitySelector' in opts and opts['DensitySelector'] is not None:
from eden.selector import DensitySelector
selectors.append(DensitySelector(n_instances=opts['DensitySelector']))
if 'SparseSelector' in opts and opts['SparseSelector'] is not None:
from eden.selector import SparseSelector
selectors.append(SparseSelector(n_instances=opts['SparseSelector']))
if 'EqualizingSelector' in opts and opts['EqualizingSelector'] is not None:
from sklearn.cluster import MiniBatchKMeans
from eden.selector import EqualizingSelector
n_clusters = opts['EqualizingSelector'] / 10
selectors.append(
EqualizingSelector(n_instances=opts['EqualizingSelector'],
clustering_algo=MiniBatchKMeans(n_clusters)))
return selectors
def __init__(self,
compiled=False,
learning_rate=0.002,
n_layers=1,
n_features_hidden_factor=10,
selectors=[QuickShiftSelector()],
n_nearest_neighbors=10,
n_links=1,
layout='force',
layout_prog='sfdp',
layout_prog_args='-Goverlap=scale',
n_eigenvectors=10,
random_state=1,
metric='rbf',
**kwds):
self.compiled = compiled
self.learning_rate = learning_rate
self.n_layers = n_layers
self.n_features_hidden_factor = n_features_hidden_factor
self.selectors = selectors
self.n_nearest_neighbors = n_nearest_neighbors
self.n_links = n_links
self.layout = layout
self.layout_prog = layout_prog
self.layout_prog_args = layout_prog_args
self.n_eigenvectors = n_eigenvectors
self.metric = metric
self.kwds = kwds
self.random_state = random_state
self.selected_instances_list = []
self.selected_instances_ids_list = []