def evaluateDG( fgold, fmodel, ignoreroot=True, minlength=1 ):
m = Metrics(ignoreroot)
for dggold, dgmodel in zip(idg(fgold), idg(fmodel)):
if dggold.length() >= minlength:
if not dggold.__eq__(dgmodel):
warnings.warn("Two sentences are not equal %s %s" % (str(dggold), str(dgmodel)))
m.add(dggold, dgmodel)
print str(m)
def _file_content(self, filename): nsentence = 0 lengths = [] for dg in idg(filename): nsentence+=1 lengths.append(dg.length()) return nsentence, lengths
def test_dependency_graph(self):
from sets import Set
for i, dg in enumerate(idg('conll/00/wsj_0001.dp')):
if i == 0:
self.assertEqual(Set([(2, 7), (12, 15), (5, 4), (2, 6), (15, 13), (8, 2), (11, 10), (2, 1), (8, 9), (9, 16), (9, 11), (2, 3), (9, 12), (8, 18), (15, 14), (0, 8), (6, 5), (16, 17)]), dg.edgeset(False))
self.assertEqual(Set([(2, 7), (12, 15), (5, 4), (2, 6), (15, 13), (8, 2), (11, 10), (2, 1), (8, 9), (9, 16), (9, 11), (2, 3), (9, 12), (8, 18), (15, 14), (6, 5), (16, 17)]), dg.edgeset(True))
else:
self.assertEqual(Set([(2,1), (3,2), (0,3),(3,4),(4,5),(5,6),(6,7),(6,8),(12,9),(12,10),(12,11),(6,12),(3,13)]), dg.edgeset(False))
self.assertEqual(Set([(2,1), (3,2),(3,4),(4,5),(5,6),(6,7),(6,8),(12,9),(12,10),(12,11),(6,12),(3,13)]), dg.edgeset(True))
def test_wsj10_sentence_length_freq(self):
expected = {1: 159, 2: 340, 3: 377, 4: 518, 5: 614, 6: 737, 7: 878, 8: 1107, 9: 1208, 10: 1484}
actual = {}
for dg in idg(wsj10file):
if dg.length() not in actual:
actual[dg.length()] = 0
actual[dg.length()]+=1
self.assertEqual(expected, actual)
def test_rhead_metrics(self):
from parser import rhead
from common import Metrics
m_ignoreroot,m = Metrics(True),Metrics(False)
for dg in idg('conll/00/wsj_0001.dp'):
m_ignoreroot.add(dg, rhead(dg))
m.add(dg, rhead(dg))
self.assertEqual(31, m.directed.nreference)
self.assertEqual(29, m_ignoreroot.directed.nreference)
self.assertEqual(15, m_ignoreroot.undirected.nmatch)
self.assertEqual(8, m_ignoreroot.directed.nmatch)
self.assertEqual(15, m.undirected.nmatch)
self.assertEqual(8, m.directed.nmatch)
def test_wsj10_postag_freq(self):
expected = {'PRP$': 412, 'VBG': 735, 'VBD': 2633, 'VBN': 1282, 'VBP': 1361, 'WDT': 66,
'JJ': 3658, 'WP': 145, 'VBZ': 2320, 'DT': 4586, 'RP': 141, 'NN': 7718, 'FW': 22, 'POS': 332, 'TO': 1183, 'PRP': 2000, 'RB': 3071,
'NNS': 3927, 'NNP': 5570, 'VB': 1616, 'WRB': 96, 'CC': 1036, 'LS': 24, 'PDT': 31, 'RBS': 26, 'RBR': 113, 'CD': 3004,
'EX': 120, 'IN': 3720, 'MD': 678, 'NNPS': 192, 'JJS': 106, 'JJR': 228, 'SYM': 51, 'UH': 45}
actual = {}
for dg in idg(wsj10file):
for n in dg.nodeiter():
if n.postag() not in actual:
actual[n.postag()] = 0
actual[n.postag()]+=1
self.assertEqual(expected, actual)
def tesit_dg10_sentence_count(self): self.assertEqual(7422, len([dg for dg in idg(wsj10file)]))
