def sample_data(n=200):
tagged_corpus = TaggedCorpus()
file = join(dirname(__file__), "corpus", "vlsp_chunk", "train.txt")
tagged_corpus.load(file)
sentences = tagged_corpus.sentences[:n]
sample_corpus = TaggedCorpus(sentences)
file = join(dirname(__file__), "corpus", "vlsp_chunk_sample", "train.txt")
sample_corpus.save(file)
def raw_to_corpus():
for f in ["train.txt", "dev.txt", "test.txt"]:
tagged_corpus = TaggedCorpus()
input = join(dirname(dirname(__file__)), "raw", "vlsp2016", f)
tagged_corpus.load(input)
tagged_corpus.sentences = preprocess(tagged_corpus.sentences)
output = join(dirname(dirname(__file__)), "corpus", "vlsp2016", f)
tagged_corpus.save(output)
def raw_to_sample_corpus(n):
for f in ["train.txt", "dev.txt", "test.txt"]:
tagged_corpus = TaggedCorpus()
input = join(dirname(dirname(__file__)), "raw", "treebank", f)
tagged_corpus.load(input)
tagged_corpus.sentences = preprocess(tagged_corpus.sentences)[:n]
output = join(dirname(dirname(__file__)), "corpus", "sample_treebank",
f)
tagged_corpus.save(output)
def raw_to_corpus():
for f1, f2 in [("chunk_train.txt", "train.txt"),
("chunk_dev.txt", "dev.txt"),
("chunk_test.txt", "test.txt")]:
tagged_corpus = TaggedCorpus()
input = join(dirname(dirname(__file__)), "raw", "treebank", f1)
tagged_corpus.load(input)
tagged_corpus.sentences = preprocess(tagged_corpus.sentences)
output = join(dirname(dirname(__file__)), "corpus", "treebank", f2)
tagged_corpus.save(output)