Upload py2hz.py
Browse files
py2hz.py
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
# _*_ coding:utf-8 _*_
|
2 |
"""
|
3 |
-
@Version : 1.
|
4 |
-
@Time : 2024年12月
|
5 |
@Author : DuYu (@duyu09, [email protected])
|
6 |
@File : py2hz.py
|
7 |
@Describe : 基于隐马尔可夫模型(HMM)的拼音转汉字程序。
|
@@ -66,6 +66,8 @@ def train_hmm(sentences, pinyins, hanzi2id, pinyin2id):
|
|
66 |
pinyin_seq = [pinyin2id[p] for p in pinyin.split()]
|
67 |
|
68 |
# 初始状态概率
|
|
|
|
|
69 |
start_prob[hanzi_seq[0]] += 1
|
70 |
|
71 |
# 转移概率
|
@@ -140,5 +142,5 @@ def pred(model_path='hmm_model.pkl.bz2', pinyin_str='ce4 shi4'):
|
|
140 |
print('预测结果:', result)
|
141 |
|
142 |
if __name__ == '__main__':
|
143 |
-
# train(dataset_path='
|
144 |
pred(model_path='hmm_model.pkl.bz2', pinyin_str='hong2 yan2 bo2 ming4')
|
|
|
1 |
# _*_ coding:utf-8 _*_
|
2 |
"""
|
3 |
+
@Version : 1.1.0
|
4 |
+
@Time : 2024年12月28日
|
5 |
@Author : DuYu (@duyu09, [email protected])
|
6 |
@File : py2hz.py
|
7 |
@Describe : 基于隐马尔可夫模型(HMM)的拼音转汉字程序。
|
|
|
66 |
pinyin_seq = [pinyin2id[p] for p in pinyin.split()]
|
67 |
|
68 |
# 初始状态概率
|
69 |
+
if len(hanzi_seq) == 0:
|
70 |
+
continue
|
71 |
start_prob[hanzi_seq[0]] += 1
|
72 |
|
73 |
# 转移概率
|
|
|
142 |
print('预测结果:', result)
|
143 |
|
144 |
if __name__ == '__main__':
|
145 |
+
# train(dataset_path='train_o.csv', model_path='hmm_model.pkl.bz2')
|
146 |
pred(model_path='hmm_model.pkl.bz2', pinyin_str='hong2 yan2 bo2 ming4')
|