def _update_labels(self):
labels = []
if self.root and self._displayed_root:
indices = [leaf.value.index for leaf in leaves(self.root)]
if self.annotation_idx == 0:
labels = []
elif self.annotation_idx == 1:
labels = [str(i+1) for i in indices]
elif self.label_cb.model()[self.annotation_idx] == "Attribute names":
attr = self.matrix.row_items.domain.attributes
labels = [str(attr[i]) for i in indices]
elif isinstance(self.items, Orange.data.Table):
var = self.label_cb.model()[self.annotation_idx]
col_data, _ = self.items.get_column_view(var)
labels = [var.str_val(val) for val in col_data]
labels = [labels[idx] for idx in indices]
else:
labels = []
if labels and self._displayed_root is not self.root:
joined = leaves(self._displayed_root)
labels = [", ".join(labels[leaf.value.first: leaf.value.last])
for leaf in joined]
self.labels.set_labels(labels)
self.labels.setMinimumWidth(1 if labels else -1)
def _update_labels(self):
labels = []
if self.root and self._displayed_root:
indices = [leaf.value.index for leaf in leaves(self.root)]
if self.annotation_idx == 0:
labels = []
elif self.annotation_idx == 1:
labels = [str(i) for i in indices]
elif isinstance(self.items, Orange.data.Table):
var = self.label_cb.model()[self.annotation_idx]
col = self.items[:, var]
labels = [var.repr_val(next(iter(row))) for row in col]
labels = [labels[idx] for idx in indices]
else:
labels = []
if labels and self._displayed_root is not self.root:
joined = leaves(self._displayed_root)
labels = [
", ".join(labels[leaf.value.first:leaf.value.last])
for leaf in joined
]
self.labels.set_labels(labels)
self.labels.setMinimumWidth(1 if labels else -1)
def _update_labels(self):
labels = []
if self.root and self._displayed_root:
indices = [leaf.value.index for leaf in leaves(self.root)]
if self.annotation_idx == 0:
labels = []
elif self.annotation_idx == 1:
labels = [str(i) for i in indices]
elif self.label_cb.model()[
self.annotation_idx] == "Attribute names":
attr = self.matrix.row_items.domain.attributes
labels = [str(attr[i]) for i in indices]
elif isinstance(self.items, Orange.data.Table):
var = self.label_cb.model()[self.annotation_idx]
col_data, _ = self.items.get_column_view(var)
labels = [var.str_val(val) for val in col_data]
labels = [labels[idx] for idx in indices]
else:
labels = []
if labels and self._displayed_root is not self.root:
joined = leaves(self._displayed_root)
labels = [
", ".join(labels[leaf.value.first:leaf.value.last])
for leaf in joined
]
self.labels.set_labels(labels)
self.labels.setMinimumWidth(1 if labels else -1)
def _update_labels(self):
labels = []
if self.root and self._displayed_root:
indices = [leaf.value.index for leaf in leaves(self.root)]
if self.annotation == "None":
labels = []
elif self.annotation == "Enumeration":
labels = [str(i + 1) for i in indices]
elif self.annotation == "Name":
attr = self.matrix.row_items.domain.attributes
labels = [str(attr[i]) for i in indices]
elif isinstance(self.annotation, Orange.data.Variable):
col_data, _ = self.items.get_column_view(self.annotation)
labels = [self.annotation.str_val(val) for val in col_data]
labels = [labels[idx] for idx in indices]
else:
labels = []
if labels and self._displayed_root is not self.root:
joined = leaves(self._displayed_root)
labels = [
", ".join(labels[leaf.value.first:leaf.value.last])
for leaf in joined
]
self.labels.setItems(labels)
self.labels.setMinimumWidth(1 if labels else -1)
def leaf_items(self):
"""Iterate over the dendrogram leaf items (:class:`QGraphicsItem`).
"""
if self._root:
return (self._items[leaf] for leaf in leaves(self._root))
else:
return iter(())
def leaf_items(self):
"""Iterate over the dendrogram leaf items (:class:`QGraphicsItem`).
"""
if self._root:
return (self._items[leaf] for leaf in leaves(self._root))
else:
return iter(())
def test_mapping(self):
leaves = list(hierarchical.leaves(self.cluster))
indices = [n.value.index for n in leaves]
self.assertEqual(len(indices), len(self.matrix.items))
self.assertEqual(set(indices), set(range(len(self.matrix.items))))
self.assertEqual(indices, [3, 1, 2, 6, 0, 4, 8, 9, 5, 7])
def test_mapping(self):
leaves = list(hierarchical.leaves(self.cluster))
indices = [n.value.index for n in leaves]
self.assertEqual(len(indices), len(self.matrix.items))
self.assertEqual(set(indices), set(range(len(self.matrix.items))))
self.assertEqual(indices,
[3, 1, 2, 6, 0, 4, 8, 9, 5, 7])
def cluster_data(self, matrix):
with self.progressBar():
# cluster rows
if len(matrix) > 1:
rows_distances = Euclidean(matrix)
cluster = hierarchical.dist_matrix_clustering(rows_distances)
row_order = hierarchical.optimal_leaf_ordering(
cluster, rows_distances, progress_callback=self.progressBarSet)
row_order = np.array([x.value.index for x in leaves(row_order)])
else:
row_order = np.array([0])
# cluster columns
if matrix.X.shape[1] > 1:
columns_distances = Euclidean(matrix, axis=0)
cluster = hierarchical.dist_matrix_clustering(columns_distances)
columns_order = hierarchical.optimal_leaf_ordering(
cluster, columns_distances,
progress_callback=self.progressBarSet)
columns_order = np.array([x.value.index for x in leaves(columns_order)])
else:
columns_order = np.array([0])
return row_order, columns_order
def _restore_selection(self, state):
# type: (SelectionState) -> bool
"""
Restore the (manual) node selection state.
Return True if successful; False otherwise.
"""
linkmatrix = self.linkmatrix
if self.selection_method == 0 and self.root:
selected, linksaved = state
linkstruct = np.array(linksaved, dtype=float)
selected = set(selected) # type: Set[Tuple[int]]
if not selected:
return False
if linkmatrix.shape[0] != linkstruct.shape[0]:
return False
# check that the linkage matrix structure matches. Use isclose for
# the height column to account for inexact floating point math
# (e.g. summation order in different ?gemm implementations for
# euclidean distances, ...)
if np.any(linkstruct[:, :2] != linkmatrix[:, :2]) or \
not np.all(np.isclose(linkstruct[:, 2], linkstruct[:, 2])):
return False
selection = []
indices = np.array([n.value.index for n in leaves(self.root)],
dtype=int)
# mapping from ranges to display (pruned) nodes
mapping = {
node.value.range: node
for node in postorder(self._displayed_root)
}
for node in postorder(self.root): # type: Tree
r = tuple(indices[node.value.first:node.value.last])
if r in selected:
if node.value.range not in mapping:
# the node was pruned from display and cannot be
# selected
break
selection.append(mapping[node.value.range])
selected.remove(r)
if not selected:
break # found all, nothing more to do
if selection and selected:
# Could not restore all selected nodes (only partial match)
return False
self._set_selected_nodes(selection)
return True
return False
def _update_labels(self):
labels = []
if self.root and self._displayed_root:
indices = [leaf.value.index for leaf in leaves(self.root)]
if self.annotation_idx == 0:
labels = []
elif self.annotation_idx == 1:
labels = [str(i) for i in indices]
elif isinstance(self.items, Orange.data.Table):
var = self.label_cb.model()[self.annotation_idx]
col = self.items[:, var]
labels = [var.repr_val(next(iter(row))) for row in col]
labels = [labels[idx] for idx in indices]
else:
labels = []
if labels and self._displayed_root is not self.root:
joined = leaves(self._displayed_root)
labels = [", ".join(labels[leaf.value.first: leaf.value.last])
for leaf in joined]
self.labels.set_labels(labels)
self.labels.setMinimumWidth(1 if labels else -1)
def _rescale(self):
if self._root is None:
return
crect = self.contentsRect()
leaf_count = len(list(leaves(self._root)))
if self.orientation in [Left, Right]:
drect = QSizeF(self._root.value.height, leaf_count - 1)
else:
drect = QSizeF(self._root.value.last - 1, self._root.value.height)
transform = QTransform().scale(crect.width() / drect.width(), crect.height() / drect.height())
self._itemgroup.setPos(crect.topLeft())
self._itemgroup.setTransform(transform)
self._selection_items = None
self._update_selection_items()
def _update_ordering(self):
if self.sorting == OWDistanceMap.NoOrdering:
self._sorted_matrix = self.matrix
self._sort_indices = None
else:
if self.sorting == OWDistanceMap.Clustering:
tree = self._cluster_tree()
elif self.sorting == OWDistanceMap.OrderedClustering:
tree = self._ordered_cluster_tree()
leaves = hierarchical.leaves(tree)
indices = numpy.array([leaf.value.index for leaf in leaves])
X = self.matrix
self._sorted_matrix = X[indices[:, numpy.newaxis],
indices[numpy.newaxis, :]]
self._sort_indices = indices
def _save_selection(self):
# Save the current manual node selection state
selection_state = None
if self.selection_method == 0 and self.root:
assert self.linkmatrix is not None
linkmat = [(int(_0), int(_1), _2)
for _0, _1, _2 in self.linkmatrix[:, :3].tolist()]
nodes_ = self.dendrogram.selected_nodes()
# match the display (pruned) nodes back (by ranges)
mapping = {node.value.range: node for node in postorder(self.root)}
nodes = [mapping[node.value.range] for node in nodes_]
indices = [tuple(node.value.index for node in leaves(node))
for node in nodes]
if nodes:
selection_state = (indices, linkmat)
return selection_state
def _rescale(self):
if self._root is None:
return
crect = self.contentsRect()
leaf_count = len(list(leaves(self._root)))
if self.orientation in [Left, Right]:
drect = QSizeF(self._root.value.height, leaf_count - 1)
else:
drect = QSizeF(self._root.value.last - 1, self._root.value.height)
transform = QTransform().scale(crect.width() / drect.width(),
crect.height() / drect.height())
self._itemgroup.setPos(crect.topLeft())
self._itemgroup.setTransform(transform)
self._selection_items = None
self._update_selection_items()
def _rescale(self):
if self._root is None:
return
scale = self._height_scale_factor()
base = scale * self._root.value.height
crect = self.contentsRect()
leaf_count = len(list(leaves(self._root)))
if self.orientation in [Left, Right]:
drect = QSizeF(base, leaf_count)
else:
drect = QSizeF(leaf_count, base)
eps = np.finfo(np.float64).eps
if abs(drect.width()) < eps:
sx = 1.0
else:
sx = crect.width() / drect.width()
if abs(drect.height()) < eps:
sy = 1.0
else:
sy = crect.height() / drect.height()
transform = QTransform().scale(sx, sy)
self._transform = transform
self._itemgroup.setPos(crect.topLeft())
self._itemgroup.setGeometry(crect)
for node_geom in postorder(self._layout):
node, _ = node_geom.value
item = self._items[node]
item.setGeometryData(transform.map(item.sourcePath),
transform.map(item.sourceAreaShape))
self._selection_items = None
self._update_selection_items()
def indices(root):
return [leaf.value.index for leaf in hierarchical.leaves(root)]
def commit(self):
items = getattr(self.matrix, "items", self.items)
if not items:
# nothing to commit
return
selection = self.dendrogram.selected_nodes()
selection = sorted(selection, key=lambda c: c.value.first)
indices = [leaf.value.index for leaf in leaves(self.root)]
maps = [indices[node.value.first:node.value.last]
for node in selection]
selected_indices = list(chain(*maps))
unselected_indices = sorted(set(range(self.root.value.last)) -
set(selected_indices))
selected = [items[k] for k in selected_indices]
unselected = [items[k] for k in unselected_indices]
if not selected:
self.send("Selected Data", None)
self.send("Other Data", None)
return
selected_data = unselected_data = None
if isinstance(items, Orange.data.Table):
c = numpy.zeros(len(items))
for i, indices in enumerate(maps):
c[indices] = i
c[unselected_indices] = len(maps)
mask = c != len(maps)
if self.append_clusters:
clust_var = Orange.data.DiscreteVariable(
str(self.cluster_name),
values=["Cluster {}".format(i + 1)
for i in range(len(maps))] +
["Other"], ordered=True
)
data, domain = items, items.domain
attrs = domain.attributes
class_ = domain.class_vars
metas = domain.metas
if self.cluster_role == self.AttributeRole:
attrs = attrs + (clust_var,)
elif self.cluster_role == self.ClassRole:
class_ = class_ + (clust_var,)
elif self.cluster_role == self.MetaRole:
metas = metas + (clust_var,)
domain = Orange.data.Domain(attrs, class_, metas)
data = Orange.data.Table(domain, data)
data.get_column_view(clust_var)[0][:] = c
else:
data = items
if selected:
selected_data = data[mask]
if unselected:
unselected_data = data[~mask]
self.send("Selected Data", selected_data)
self.send("Other Data", unselected_data)
def commit(self):
items = getattr(self.matrix, "items", self.items)
if not items:
# nothing to commit
return
selection = self.dendrogram.selected_nodes()
selection = sorted(selection, key=lambda c: c.value.first)
indices = [leaf.value.index for leaf in leaves(self.root)]
maps = [
indices[node.value.first:node.value.last] for node in selection
]
selected_indices = list(chain(*maps))
unselected_indices = sorted(
set(range(self.root.value.last)) - set(selected_indices))
if not selected_indices:
self.send("Selected Data", None)
self.send("Other Data", None)
return
selected_data = unselected_data = None
if isinstance(items, Orange.data.Table) and self.matrix.axis == 1:
# Select rows
c = numpy.zeros(self.matrix.X.shape[0])
for i, indices in enumerate(maps):
c[indices] = i
c[unselected_indices] = len(maps)
mask = c != len(maps)
if self.append_clusters:
clust_var = Orange.data.DiscreteVariable(
str(self.cluster_name),
values=[
"Cluster {}".format(i + 1) for i in range(len(maps))
] + ["Other"])
data, domain = items, items.domain
attrs = domain.attributes
class_ = domain.class_vars
metas = domain.metas
if self.cluster_role == self.AttributeRole:
attrs = attrs + (clust_var, )
elif self.cluster_role == self.ClassRole:
class_ = class_ + (clust_var, )
elif self.cluster_role == self.MetaRole:
metas = metas + (clust_var, )
domain = Orange.data.Domain(attrs, class_, metas)
data = Orange.data.Table.from_table(domain, items)
data.get_column_view(clust_var)[0][:] = c
else:
data = items
if selected_indices:
selected_data = data[mask]
if unselected_indices:
unselected_data = data[~mask]
elif isinstance(items, Orange.data.Table) and self.matrix.axis == 0:
# Select columns
domain = Orange.data.Domain(
[items.domain[i] for i in selected_indices],
items.domain.class_vars, items.domain.metas)
selected_data = items.from_table(domain, items)
domain = Orange.data.Domain(
[items.domain[i] for i in unselected_indices],
items.domain.class_vars, items.domain.metas)
unselected_data = items.from_table(domain, items)
self.send("Selected Data", selected_data)
self.send("Other Data", unselected_data)
def indices(root):
return [leaf.value.index for leaf in hierarchical.leaves(root)]
def commit(self):
items = getattr(self.matrix, "items", self.items)
if not items:
self.Outputs.selected_data.send(None)
self.Outputs.annotated_data.send(None)
return
selection = self.dendrogram.selected_nodes()
selection = sorted(selection, key=lambda c: c.value.first)
indices = [leaf.value.index for leaf in leaves(self.root)]
maps = [
indices[node.value.first:node.value.last] for node in selection
]
selected_indices = list(chain(*maps))
unselected_indices = sorted(
set(range(self.root.value.last)) - set(selected_indices))
if not selected_indices:
self.Outputs.selected_data.send(None)
annotated_data = create_annotated_table(items, []) \
if self.selection_method == 0 and self.matrix.axis else None
self.Outputs.annotated_data.send(annotated_data)
return
selected_data = None
if isinstance(items, Orange.data.Table) and self.matrix.axis == 1:
# Select rows
c = np.zeros(self.matrix.shape[0])
for i, indices in enumerate(maps):
c[indices] = i
c[unselected_indices] = len(maps)
mask = c != len(maps)
data, domain = items, items.domain
attrs = domain.attributes
classes = domain.class_vars
metas = domain.metas
var_name = get_unique_names(domain, "Cluster")
values = [f"C{i + 1}" for i in range(len(maps))]
clust_var = Orange.data.DiscreteVariable(var_name,
values=values + ["Other"])
domain = Orange.data.Domain(attrs, classes, metas + (clust_var, ))
data = items.transform(domain)
with data.unlocked(data.metas):
data.get_column_view(clust_var)[0][:] = c
if selected_indices:
selected_data = data[mask]
clust_var = Orange.data.DiscreteVariable(var_name,
values=values)
selected_data.domain = Domain(attrs, classes,
metas + (clust_var, ))
annotated_data = create_annotated_table(data, selected_indices)
elif isinstance(items, Orange.data.Table) and self.matrix.axis == 0:
# Select columns
attrs = []
for clust, indices in chain(enumerate(maps, start=1),
[(0, unselected_indices)]):
for i in indices:
attr = items.domain[i].copy()
attr.attributes["cluster"] = clust
attrs.append(attr)
domain = Orange.data.Domain(
# len(unselected_indices) can be 0
attrs[:len(attrs) - len(unselected_indices)],
items.domain.class_vars,
items.domain.metas)
selected_data = items.from_table(domain, items)
domain = Orange.data.Domain(attrs, items.domain.class_vars,
items.domain.metas)
annotated_data = items.from_table(domain, items)
self.Outputs.selected_data.send(selected_data)
self.Outputs.annotated_data.send(annotated_data)
def commit(self):
self._invalidated = False
items = getattr(self.matrix, "items", self.items)
if not items:
# nothing to commit
return
selection = self.dendrogram.selected_nodes()
selection = sorted(selection, key=lambda c: c.value.first)
indices = [leaf.value.index for leaf in leaves(self.root)]
maps = [
indices[node.value.first:node.value.last] for node in selection
]
selected_indices = list(chain(*maps))
unselected_indices = sorted(
set(range(self.root.value.last)) - set(selected_indices))
selected = [items[k] for k in selected_indices]
unselected = [items[k] for k in unselected_indices]
if not selected:
self.send("Selected Data", None)
self.send("Other Data", None)
return
selected_data = unselected_data = None
if isinstance(items, Orange.data.Table):
c = numpy.zeros(len(items))
for i, indices in enumerate(maps):
c[indices] = i
c[unselected_indices] = len(maps)
mask = c != len(maps)
if self.append_clusters:
clust_var = Orange.data.DiscreteVariable(
str(self.cluster_name),
values=[
"Cluster {}".format(i + 1) for i in range(len(maps))
] + ["Other"])
data, domain = items, items.domain
attrs = domain.attributes
class_ = domain.class_vars
metas = domain.metas
X, Y, M = data.X, data.Y, data.metas
if self.cluster_role == self.AttributeRole:
attrs = attrs + (clust_var, )
X = numpy.c_[X, c]
elif self.cluster_role == self.ClassRole:
class_ = class_ + (clust_var, )
Y = numpy.c_[Y, c]
elif self.cluster_role == self.MetaRole:
metas = metas + (clust_var, )
M = numpy.c_[M, c]
domain = Orange.data.Domain(attrs, class_, metas)
data = Orange.data.Table(domain, X, Y, M)
else:
data = items
if selected:
selected_data = data[mask]
if unselected:
unselected_data = data[~mask]
self.send("Selected Data", selected_data)
self.send("Other Data", unselected_data)
