-
Notifications
You must be signed in to change notification settings - Fork 0
/
run_experiments.py
38 lines (30 loc) · 1.29 KB
/
run_experiments.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
from slang.corpus_processing import CorpusProcessing
from slang.trainer import Trainer
def main():
# CoNLL2003
connl2003 = CorpusProcessing(corpus='connl2003')
connl2003.process("datasets/CoNLL2003/")
# connl2003.load_embeddings('glove.6B.100d.txt', emb_type='glove')
trainer = Trainer(connl2003)
trainer.train_and_evaluate(epochs=2)
# ToDo: generate data on the fly, memory consumption
# https://stanford.edu/~shervine/blog/keras-how-to-generate-data-on-the-fly
# Paramopama
# paramopama = CorpusProcessing(corpus='paramopama')
# paramopama.process("datasets/Paramopama/")
# # paramopama.convert_tags()
# paramopama.split_corpus(split=0.6)
# paramopama.load_embeddings('embeddings/CHAVE/CHAVE_word2vec.keyed', emb_type='word2vec')
# #paramopama.load_embeddings('embeddings/publico_vectors_non-breaking-spaces.bin', emb_type='word2vec')
# trainer = Trainer(paramopama)
# trainer.train_and_evaluate(epochs=1)
# CINTIL
# cintil = CorpusProcessing(corpus='cintil')
# cintil.process("datasets/CINTIL/")
# cintil.split_corpus(split=0.8)
# trainer = Trainer(cintil)
# trainer.train_and_evaluate(epochs=1)
# trainer = Trainer(comtravo)
# trainer.cross_fold_evaluation(epochs=5)
if __name__ == "__main__":
main()