def decode(self, s_arc, mask):
lens = mask.sum(1)
# prevent self-loops
s_arc.diagonal(0, 1, 2).fill_(float('-inf'))
arc_preds = mst(s_arc, mask)
return arc_preds
def parse(self,
pretrained_word_tokens=None,
word_tokens=None,
pos_tokens=None):
if word_tokens is not None:
self.forward(pretrained_word_tokens, word_tokens, pos_tokens)
ROOT = self._ROOT_LABEL
arcs_batch, labels_batch = [], []
arc_logits = self._arc_logits.data.cpu().numpy()
label_logits = self._label_logits.data.cpu().numpy()
for arc_logit, label_logit, length in \
zip(arc_logits, np.transpose(label_logits, (0, 2, 1, 3)),
self._lengths):
arc_probs = softmax2d(arc_logit[:length, :length])
arcs = mst(arc_probs)
label_probs = softmax2d(label_logit[np.arange(length), arcs])
labels = np.argmax(label_probs, axis=1)
labels[0] = ROOT
tokens = np.arange(1, length)
roots = np.where(labels[tokens] == ROOT)[0] + 1
if len(roots) < 1:
root_arc = np.where(arcs[tokens] == 0)[0] + 1
labels[root_arc] = ROOT
elif len(roots) > 1:
label_probs[roots, ROOT] = 0
new_labels = \
np.argmax(label_probs[roots], axis=1)
root_arc = np.where(arcs[tokens] == 0)[0] + 1
labels[roots] = new_labels
labels[root_arc] = ROOT
arcs_batch.append(arcs)
labels_batch.append(labels)
return arcs_batch, labels_batch
def __init__(self, arr, genotype=None):
self.rows = arr.shape[0]
self.cols = arr.shape[1]
# Generate initial phenotype
if genotype is None:
self.genotype = mst(arr, self.rows, self.cols)
else:
self.genotype = genotype
self.segments = find_segments(self.genotype)
self.edge_value = edge_value(arr, self.segments, self.rows, self.cols)
self.connectivity = connectivity(arr, self.segments, self.rows,
self.cols)
self.deviation = overall_deviation(arr, self.segments, self.cols)
self.fitness = 10 * self.edge_value - 8500 * self.connectivity - 2 * self.deviation
def __init__(self, arr, genotype=None):
self.rows = arr.shape[0]
self.cols = arr.shape[1]
# Generate initial phenotype
if genotype is None:
self.genotype = mst(arr, self.rows, self.cols)
else:
self.genotype = genotype
self.segments = find_segments(self.genotype)
self.num_segments = max(self.segments) + 1
self.edge_value = -edge_value(arr, self.segments, self.rows,
self.cols) # negative to minimize
self.connectivity = connectivity(arr, self.segments, self.rows,
self.cols)
self.deviation = overall_deviation(arr, self.segments, self.cols)