def token_type(disc_clsdict, wrd_corpus, fragments_cross,
dest, verbose, n_jobs):
if verbose:
print banner('TOKEN/TYPE')
ptoc, rtoc, ptyc, rtyc = _token_type_sub(disc_clsdict, wrd_corpus,
fragments_cross, 'cross',
verbose, n_jobs)
ftoc = np.fromiter((fscore(ptoc[i], rtoc[i]) for i in xrange(ptoc.shape[0])),
dtype=np.double)
ftyc = np.fromiter((fscore(ptyc[i], rtyc[i]) for i in xrange(ptyc.shape[0])),
dtype=np.double)
with open(path.join(dest, 'token_type'), 'w') as fid:
fid.write(pretty_score(ptoc, rtoc, ftoc, 'token total',
sum(map(len, fragments_cross))))
fid.write('\n')
fid.write(pretty_score(ptyc, rtyc, ftyc, 'type total',
sum(map(len, fragments_cross))))
def boundary(disc_clsdict, corpus, fragments_cross,
dest, verbose, n_jobs):
if verbose:
print banner('BOUNDARY')
pc, rc = _boundary_sub(disc_clsdict, corpus, fragments_cross,
'cross', verbose, n_jobs)
fc = np.fromiter((fscore(pc[i], rc[i]) for i in xrange(pc.shape[0])), dtype=np.double)
with open(path.join(dest, 'boundary'), 'w') as fid:
fid.write(pretty_score(pc, rc, fc, 'boundary total',
sum(map(len, fragments_cross))))
def group(disc_clsdict, fragments_all, dest, verbose, n_jobs):
if verbose:
print banner('GROUP')
#TODO CHECK SCORE ACROSS/WITHIN!
pc, rc = _group_sub(disc_clsdict, fragments_all, 'all', verbose, n_jobs)
fc = np.fromiter((fscore(pc[i], rc[i]) for i in xrange(pc.shape[0])), dtype=np.double)
#pw, rw = _group_sub(disc_clsdict, fragments_within, 'within', verbose, n_jobs)
#fw = np.fromiter((fscore(pw[i], rw[i]) for i in xrange(pw.shape[0])), dtype=np.double)
with open(path.join(dest, 'group'), 'w') as fid:
fid.write(pretty_score(pc, rc, fc, 'group total',
sum(map(len, fragments_all))))