def main(path, draw=True):
with gzip.open("resources/bc3200.pickle.gz") as fin:
print 'Load Brown clusters for creating features ...'
bcvocab = load(fin)
evalparser(path=path, report=False, draw=draw,
bcvocab=bcvocab,
withdp=False)
def main(path, draw=True):
with gzip.open("resources/bc3200.pickle.gz") as fin:
print('Load Brown clusters for creating features ...')
bcvocab = load(fin)
evalparser(path=path, report=False, draw=draw,
bcvocab=bcvocab,
withdp=False)
print "len(vocab) = {}".format(len(vocab))
data = Data()
trnM, trnL = data.loadmatrix(fdata, flabel)
print "trnM.shape = {}".format(trnM.shape)
idxlabelmap = reversedict(labelidxmap)
pm = ParsingModel(vocab=vocab, idxlabelmap=idxlabelmap)
pm.train(trnM, trnL)
pm.savemodel("model/parsing-model.pickle.gz")
if __name__ == "__main__":
bcvocab = None
## Use brown clsuters
with gzip.open("resources/bc3200.pickle.gz") as fin:
print "Load Brown clusters for creating features ..."
bcvocab = load(fin)
## Create training data
# createtrndata(path="data/training/", topn=8000, bcvocab=bcvocab)
## Train model
# trainmodel()
## Evaluate model on the RST-DT test set
evalparser(
path="data/test/",
report=True,
bcvocab=bcvocab,
draw=False,
withdp=WITHDP,
fdpvocab="data/resources/word-dict.pickle.gz",
fprojmat="data/resources/projmat.pickle.gz",
)
D = load(gzip.open(fvocab))
vocab, labelidxmap = D['vocab'], D['labelidxmap']
print 'len(vocab) = {}'.format(len(vocab))
data = Data()
trnM, trnL = data.loadmatrix(fdata, flabel)
print 'trnM.shape = {}'.format(trnM.shape)
idxlabelmap = reversedict(labelidxmap)
pm = ParsingModel(vocab=vocab, idxlabelmap=idxlabelmap)
pm.train(trnM, trnL)
pm.savemodel("model/parsing-model.pickle.gz")
if __name__ == '__main__':
bcvocab = None
## Use brown clsuters
with gzip.open("resources/bc3200.pickle.gz") as fin:
print 'Load Brown clusters for creating features ...'
bcvocab = load(fin)
## Create training data
# createtrndata(path="data/training/", topn=8000, bcvocab=bcvocab)
## Train model
# trainmodel()
## Evaluate model on the RST-DT test set
evalparser(path="data/test/",
report=True,
bcvocab=bcvocab,
draw=False,
withdp=WITHDP,
fdpvocab="data/resources/word-dict.pickle.gz",
fprojmat="data/resources/projmat.pickle.gz")
