Port over PR #1055

This commit is contained in:
ines 2017-05-13 03:25:27 +02:00
parent ee1d35bdb0
commit 4eefb288e3

View File

@ -102,10 +102,6 @@ def read_clusters(clusters_path):
def populate_vocab(vocab, clusters, probs, oov_prob): def populate_vocab(vocab, clusters, probs, oov_prob):
# Ensure probs has entries for all words seen during clustering.
for word in clusters:
if word not in probs:
probs[word] = oov_prob
for word, prob in reversed(sorted(list(probs.items()), key=lambda item: item[1])): for word, prob in reversed(sorted(list(probs.items()), key=lambda item: item[1])):
lexeme = vocab[word] lexeme = vocab[word]
lexeme.prob = prob lexeme.prob = prob