修改model到models

reproduction/chinese_word_segment/train_context.py

@@ -3,11 +3,17 @@ from fastNLP.core.instance import Instance
 from fastNLP.core.dataset import DataSet
 from fastNLP.api.pipeline import Pipeline
 from fastNLP.api.processor import FullSpaceToHalfSpaceProcessor
-
-from reproduction.chinese_word_segment.process.cws_processor import *
-from reproduction.chinese_word_segment.process.span_converter import AlphaSpanConverter, DigitSpanConverter
+from fastNLP.api.processor import IndexerProcessor
+from reproduction.chinese_word_segment.process.cws_processor import SpeicalSpanProcessor
+from reproduction.chinese_word_segment.process.cws_processor import CWSCharSegProcessor
+from reproduction.chinese_word_segment.process.cws_processor import CWSSegAppTagProcessor
+from reproduction.chinese_word_segment.process.cws_processor import Pre2Post2BigramProcessor
+from reproduction.chinese_word_segment.process.cws_processor import VocabProcessor
+
+from reproduction.chinese_word_segment.process.span_converter import AlphaSpanConverter
+from reproduction.chinese_word_segment.process.span_converter import DigitSpanConverter
 from reproduction.chinese_word_segment.io.cws_reader import NaiveCWSReader
-
+from reproduction.chinese_word_segment.models.cws_model import CWSBiLSTMSegApp
 
 tr_filename = ''
 dev_filename = ''
@@ -60,8 +66,8 @@ bigram_proc(tr_dataset)
 char_vocab_proc(tr_dataset)
 bigram_vocab_proc(tr_dataset)
 
-char_index_proc = IndexProcessor(char_vocab_proc.get_vocab(), 'char_list')
-bigram_index_proc = IndexProcessor(bigram_vocab_proc.get_vocab(), 'bigram_list')
+char_index_proc = IndexerProcessor(char_vocab_proc.get_vocab(), 'chars_list', 'indexed_chars_list')
+bigram_index_proc = IndexerProcessor(bigram_vocab_proc.get_vocab(), 'bigrams_list','indexed_bigrams_list')
 
 char_index_proc(tr_dataset)
 bigram_index_proc(tr_dataset)
@@ -81,7 +87,8 @@ bigram_index_proc(dev_dataset)
 
 # 3. 得到数据集可以用于训练了
 # TODO pretrain的embedding是怎么解决的？
-
+cws_model = CWSBiLSTMSegApp(vocab_num, embed_dim=100, bigram_vocab_num=None, bigram_embed_dim=100, num_bigram_per_char=None,
+                 hidden_size=200, bidirectional=True, embed_drop_p=None, num_layers=1, tag_size=2)