def output(self):
for node_type, count in sorted_by_value_desc(self.counts):
if node_type == 'total': continue
print "% 15s | %d" % (node_type, count)
print "% 15s | %d" % ('total', self.counts['total'])
print
for dict_name in ('predication', 'apposition', 'modification', 'adjunction', 'coordination'):
print dict_name
for tag_string, count in sorted_by_value_desc(getattr(self, dict_name + '_kinds')):
print "% 6d | (% 2d) %s" % (count, len(tag_string.split(' ')), tag_string)
print
def output(self):
for node_type, count in sorted_by_value_desc(self.counts):
if node_type == 'total': continue
print "% 15s | %d" % (node_type, count)
print "% 15s | %d" % ('total', self.counts['total'])
print
for dict_name in ('predication', 'apposition', 'modification',
'adjunction', 'coordination'):
print dict_name
for tag_string, count in sorted_by_value_desc(
getattr(self, dict_name + '_kinds')):
print "% 6d | (% 2d) %s" % (count, len(
tag_string.split(' ')), tag_string)
print
def output(self):
print 'trace types:'
for k, v in sorted_by_value_desc(self.parents):
print '% 5d | %s' % (v, k)
print 'derivs with rcs: %d/%d=%.2f%%' % (self.rcderivs, self.total,
self.rcderivs /
float(self.total) * 100.)
def output(self):
headings = ("(%s ,) %s -> %s", "(, %s) %s -> %s", "%s (%s ,) -> %s", "%s (, %s) -> %s")
for index, heading in enumerate(headings):
env_hash = getattr(self, "e%d" % index)
print heading % ("X", "Y", "Z")
print "-" * len(heading)
examples_hash = getattr(self, "e%d_examples" % index)
for (l, r, p), f in sorted_by_value_desc(env_hash):
triple = heading % (l, r, p)
print "% 10d [%28s] %-60s %s" % (
f,
analyse(C(l), C(r), C(p)),
triple,
" ".join(examples_hash[(l, r, p)]),
)
if index == 0 and (l, r, p) in self.e3 and self.e3[(l, r, p)] <= f:
alt_triple = headings[3] % (l, r, p)
alt_freq = self.e3[(l, r, p)]
print "* % 8d%32s%-60s %s" % (alt_freq, " " * 32, alt_triple, " ".join(self.e3_examples[(l, r, p)]))
elif index == 3 and (l, r, p) in self.e0 and self.e0[(l, r, p)] <= f:
alt_triple = headings[0] % (l, r, p)
alt_freq = self.e0[(l, r, p)]
print "* % 8d%32s%-60s %s" % (alt_freq, " " * 32, alt_triple, " ".join(self.e0_examples[(l, r, p)]))
def output(self):
headings = ("(%s ,) %s -> %s",
"(, %s) %s -> %s",
"%s (%s ,) -> %s",
"%s (, %s) -> %s")
for index, heading in enumerate(headings):
env_hash = getattr(self, 'e%d' % index)
print heading % ('X', 'Y', 'Z')
print "-" * len(heading)
examples_hash = getattr(self, 'e%d_examples' % index)
for (l, r, p), f in sorted_by_value_desc(env_hash):
triple = heading % (l, r, p)
print "% 10d [%28s] %-60s %s" % (f, analyse(C(l), C(r), C(p)), triple, ' '.join(examples_hash[(l, r, p)]))
if (index == 0 and (l, r, p) in self.e3 and self.e3[(l, r, p)] <= f):
alt_triple = headings[3] % (l, r, p)
alt_freq = self.e3[(l, r, p)]
print "* % 8d%32s%-60s %s" % (alt_freq, " "*32, alt_triple,
' '.join(self.e3_examples[(l, r, p)]))
elif (index == 3 and (l, r, p) in self.e0 and self.e0[(l, r, p)] <= f):
alt_triple = headings[0] % (l, r, p)
alt_freq = self.e0[(l, r, p)]
print "* % 8d%32s%-60s %s" % (alt_freq, " "*32, alt_triple,
' '.join(self.e0_examples[(l, r, p)]))
def output(self):
with file(self.growth_fn, 'w') as f:
print >>f, '\n'.join( ' '.join(map(str, xy)) for xy in self.data_points )
with file(self.rank_fn, 'w') as f:
for k, freq in sorted_by_value_desc(self.freqs):
print >>f, freq
def output(self):
as_left = {}
as_right = {}
for (l, r, p), f in self.envs.iteritems():
if l == ",":
as_left[(l, r, p)] = f
else:
as_right[(l, r, p)] = f
print ", _ -> _"
print "--------"
for (l, r, p), f in sorted_by_value_desc(as_left):
print "% 10d [%28s] %s %20s -> %s" % (f, analyse(C(l), C(r), C(p)), l, r, p)
print "_ , -> _"
print "--------"
for (l, r, p), f in sorted_by_value_desc(as_right):
print "% 10d [%28s] %20s %s -> %s" % (f, analyse(C(l), C(r), C(p)), l, r, p)
def output(self):
as_left = {}
as_right = {}
for (l, r, p), f in self.envs.iteritems():
if l == ',':
as_left[(l, r, p)] = f
else:
as_right[(l, r, p)] = f
print ", _ -> _"
print "--------"
for (l, r, p), f in sorted_by_value_desc(as_left):
print "% 10d [%28s] %s %20s -> %s" % (f, analyse(C(l), C(r), C(p)), l, r, p)
print "_ , -> _"
print "--------"
for (l, r, p), f in sorted_by_value_desc(as_right):
print "% 10d [%28s] %20s %s -> %s" % (f, analyse(C(l), C(r), C(p)), l, r, p)
def output(self):
with file(self.growth_fn, 'w') as f:
print >> f, '\n'.join(' '.join(map(str, xy))
for xy in self.data_points)
with file(self.rank_fn, 'w') as f:
for k, freq in sorted_by_value_desc(self.freqs):
print >> f, freq
def output(self):
as_left = {}
as_right = {}
for (side, l, r, p), f in self.envs.iteritems():
if side == AnalyseAbsorption.LEFT:
as_left[ (l, r, p) ] = f
elif side == AnalyseAbsorption.RIGHT:
as_right[ (l, r, p) ] = f
print "(X ,) Y -> Z"
print "--------"
for (l, r, p), f in sorted_by_value_desc(as_left):
print "% 10d [%28s] (%s ,) %s -> %s" % (f, analyse(C(l), C(r), C(p)), l, r, p)
print "X (, Y) -> Z"
print "--------"
for (l, r, p), f in sorted_by_value_desc(as_right):
print "% 10d [%28s] %s (, %s) -> %s" % (f, analyse(C(l), C(r), C(p)), l, r, p)
def output(self):
for sig, (count, examples) in sorted_by_value_desc(self.counts):
print '% 10d | %s' % (count, sig)
print '; '.join(examples[:10])
def output(self):
for comb, freq in sorted_by_value_desc(self.counts):
print "% 10s | %d" % (comb, freq)
else:
# Split unrecognised rules by type
if r:
binary[rule_tuple] += 1
else:
unary[rule_tuple] += 1
unrecognised_rules[rule_tuple] += 1
if not has_unrecognised_rules:
with_unrecognised_rules += 1
has_unrecognised_rules = True
total += 1
print "With unrecognised rules: %d/%d=%.2f" % (with_unrecognised_rules, total, with_unrecognised_rules/float(total)*100.0)
for rule, freq in sorted_by_value_desc(unrecognised_rules):
l, r, p = rule
print "% 4d | %20s %20s -> %s" % (freq, l, r, p)
print "UCP rules: %d/%d=%.2f" % (with_ucp, total, with_ucp/float(total)*100.0)
for rule, freq in sorted_by_value_desc(ucp_rules):
l, r, p = rule
print "% 4d | %20s %20s -> %s" % (freq, l, r, p)
print "Unrecognised binary rules:"
for rule, freq in sorted_by_value_desc(binary):
l, r, p = rule
print "% 4d | %20s %20s -> %s" % (freq, l, r, p)
print "Unrecognised unary rules:"
for rule, freq in sorted_by_value_desc(unary):
def output(self):
for (l, r, p), freq in sorted_by_value_desc(self.counts):
print "%8d | %15s %-15s -> %-15s [%s]" % (freq, l, r, p, analyse(l, r, p))
def output(self):
for (cat, lex_dict) in self.examples.iteritems():
print "Category %s:" % cat
for (lex, frequency
) in sorted_by_value_desc(lex_dict)[0:self.example_count]:
print "\t% 3d | %s" % (frequency, lex)
def output(self):
for key, freq in sorted_by_value_desc(self.counts):
print '% 10d | %s' % (freq, key)
def output(self):
for (cat, lex_dict) in self.examples.iteritems():
print "Category %s:" % cat
for (lex, frequency) in sorted_by_value_desc(lex_dict)[0:self.example_count]:
print "\t% 3d | %s" % (frequency, lex)
def output(self):
with file(os.path.join(self.outdir, 'failed_freqs'), 'w') as f:
for badcat, freq in sorted_by_value_desc(self.bad_freqs):
print >>f, '% 5d | %s' % (freq, badcat)
def output(self):
for punct, f in sorted_by_value_desc(self.counts):
print "%d | %s" % (f, punct)
print "Total: %d" % self.total
def output(self):
for key, freq in sorted_by_value_desc(self.counts):
print '% 10d | %s' % (freq, key)
def output(self):
for (lch, rch, parent), freq in sorted_by_value_desc(self.envs):
print "% 10d (%s %s -> %s)" % (freq, lch, rch, parent)
def output(self):
for (comma_on_left, lch, rch, parent), freq in sorted_by_value_desc(self.envs):
print "% 10d %s (%s %s -> %s)" % (freq, "L" if comma_on_left else "R", lch, rch, parent)
def output(self):
for head, freq in sorted_by_value_desc(self.headfreqs):
print '% 4d | %s' % (freq, head)
def output(self):
for comb, freq in sorted_by_value_desc(self.counts):
print "% 10s | %d" % (comb, freq)
def output(self):
for punct, f in sorted_by_value_desc(self.counts):
print "%d | %s" % (f, punct)
print "Total: %d" % self.total
def output(self):
for comb, count in sorted_by_value_desc(self.counter):
print "% 20s | %d" % (comb, count)
if r:
binary[rule_tuple] += 1
else:
unary[rule_tuple] += 1
unrecognised_rules[rule_tuple] += 1
if not has_unrecognised_rules:
with_unrecognised_rules += 1
has_unrecognised_rules = True
total += 1
print "With unrecognised rules: %d/%d=%.2f" % (with_unrecognised_rules, total,
with_unrecognised_rules /
float(total) * 100.0)
for rule, freq in sorted_by_value_desc(unrecognised_rules):
l, r, p = rule
print "% 4d | %20s %20s -> %s" % (freq, l, r, p)
print "UCP rules: %d/%d=%.2f" % (with_ucp, total,
with_ucp / float(total) * 100.0)
for rule, freq in sorted_by_value_desc(ucp_rules):
l, r, p = rule
print "% 4d | %20s %20s -> %s" % (freq, l, r, p)
print "Unrecognised binary rules:"
for rule, freq in sorted_by_value_desc(binary):
l, r, p = rule
print "% 4d | %20s %20s -> %s" % (freq, l, r, p)
print "Unrecognised unary rules:"
def output(self):
for (p, l, r), f in sorted_by_value_desc(self.rules):
print p, l, r, f
def output(self):
for (l, r, p), freq in sorted_by_value_desc(self.counts):
print "%8d | %15s %-15s -> %-15s [%s]" % (freq, l, r, p,
analyse(l, r, p))
def output(self):
for comb, count in sorted_by_value_desc(self.counter):
print "% 20s | %d" % (comb, count)
def output(self):
for sig, (count, examples) in sorted_by_value_desc(self.counts):
print '% 10d | %s' % (count, sig)
print '; '.join(examples[:10])
def output(self):
for (cat, comb, slash_index), frequency in sorted_by_value_desc(self.freqs):
print "% 60s | %2d | %25s | %d" % (cat, slash_index, comb, frequency)
def output(self):
with file(os.path.join(self.outdir, 'failed_freqs'), 'w') as f:
for badcat, freq in sorted_by_value_desc(self.bad_freqs):
print >> f, '% 5d | %s' % (freq, badcat)
def output(self):
print 'trace types:'
for k, v in sorted_by_value_desc(self.parents):
print '% 5d | %s' % (v, k)
print 'derivs with rcs: %d/%d=%.2f%%' % (self.rcderivs, self.total, self.rcderivs/float(self.total)*100.)
def output(self):
for (p, l, r), f in sorted_by_value_desc(self.rules):
print p, l, r, f
