explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-12-25 17:36:30 +03:00
Code Issues Packages Projects Releases Wiki Activity

* Clean up unused feature templates

Browse Source
This commit is contained in:
Matthew Honnibal 2015-03-24 05:08:35 +01:00
parent 411bf377d4
commit b3157927e6
2 changed files with 1 additions and 151 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

148
spacy/syntax/_parse_features.pyx
View File

@ -296,151 +296,3 @@ trigrams = (
(N0W, N0p, N0lL, N0l2L),
(N0p, N0lL, N0l2L),
)
arc_eager = (
(S0w, S0p),
(S0w,),
(S0p,),
(N0w, N0p),
(N0w,),
(N0p,),
(N1w, N1p),
(N1w,),
(N1p,),
(N2w, N2p),
(N2w,),
(N2p,),
(S0w, S0p, N0w, N0p),
(S0w, S0p, N0w),
(S0w, N0w, N0p),
(S0w, S0p, N0p),
(S0p, N0w, N0p),
(S0w, N0w),
(S0p, N0p),
(N0p, N1p),
(N0p, N1p, N2p),
(S0p, N0p, N1p),
(S1p, S0p, N0p),
(S0p, S0lp, N0p),
(S0p, S0rp, N0p),
(S0p, N0p, N0lp),
(dist, S0w),
(dist, S0p),
(dist, N0w),
(dist, N0p),
(dist, S0w, N0w),
(dist, S0p, N0p),
(S0w, S0rv),
(S0p, S0rv),
(S0w, S0lv),
(S0p, S0lv),
(N0w, N0lv),
(N0p, N0lv),
(S1w,),
(S1p,),
(S0lw,),
(S0lp,),
(S0rw,),
(S0rp,),
(N0lw,),
(N0lp,),
(S2w,),
(S2p,),
(S0l2w,),
(S0l2p,),
(S0r2w,),
(S0r2p,),
(N0l2w,),
(N0l2p,),
(S0p, S0lp, S0l2p),
(S0p, S0rp, S0r2p),
(S0p, S1p, S2p),
(N0p, N0lp, N0l2p),
(S0L,),
(S0lL,),
(S0rL,),
(N0lL,),
(S1L,),
(S0l2L,),
(S0r2L,),
(N0l2L,),
(S0w, S0rL, S0r2L),
(S0p, S0rL, S0r2L),
(S0w, S0lL, S0l2L),
(S0p, S0lL, S0l2L),
(N0w, N0lL, N0l2L),
(N0p, N0lL, N0l2L),
)
label_sets = (
(S0w, S0lL, S0l2L),
(S0p, S0rL, S0r2L),
(S0p, S0lL, S0l2L),
(S0p, S0rL, S0r2L),
(N0w, N0lL, N0l2L),
(N0p, N0lL, N0l2L),
)
extra_labels = (
(S0p, S0lL, S0lp),
(S0p, S0lL, S0l2L),
(S0p, S0rL, S0rp),
(S0p, S0rL, S0r2L),
(S0p, S0lL, S0rL),
(S1p, S0L, S0rL),
(S1p, S0L, S0lL),
)
# Koo et al (2008) dependency features, using Brown clusters.
clusters = (
# Koo et al have (head, child) --- we have S0, N0 for both.
(S0c4, N0c4),
(S0c6, N0c6),
(S0c, N0c),
(S0p, N0c4),
(S0p, N0c6),
(S0p, N0c),
(S0c4, N0p),
(S0c6, N0p),
(S0c, N0p),
# Siblings --- right arc
(S0c4, S0rc4, N0c4),
(S0c6, S0rc6, N0c6),
(S0p, S0rc4, N0c4),
(S0c4, S0rp, N0c4),
(S0c4, S0rc4, N0p),
# Siblings --- left arc
(S0c4, N0lc4, N0c4),
(S0c6, N0c6, N0c6),
(S0c4, N0lc4, N0p),
(S0c4, N0lp, N0c4),
(S0p, N0lc4, N0c4),
# Grand-child, right-arc
(S1c4, S0c4, N0c4),
(S1c6, S0c6, N0c6),
(S1p, S0c4, N0c4),
(S1c4, S0p, N0c4),
(S1c4, S0c4, N0p),
# Grand-child, left-arc
(S0lc4, S0c4, N0c4),
(S0lc6, S0c6, N0c6),
(S0lp, S0c4, N0c4),
(S0lc4, S0p, N0c4),
(S0lc4, S0c4, N0p)
)
hasty = s0_n0 + n0_n1 + trigrams
def pos_bigrams():
kernels = [S2w, S1w, S0w, S0lw, S0rw, N0w, N0lw, N1w]
bitags = []
for t1, t2 in combinations(kernels, 2):
feat = (t1 + 1, t2 + 1)
bitags.append(feat)
print "Adding %d bitags" % len(bitags)
return tuple(bitags)

4
spacy/syntax/parser.pyx
View File

@ -58,9 +58,7 @@ cdef unicode print_state(State* s, list words):
def get_templates(name):
pf = _parse_features
if name == 'zhang':
return pf.arc_eager
elif name == 'ner':
if name == 'ner':
return pf.ner
elif name == 'debug':
return pf.unigrams