Upload 16 files
Browse files- evaluation_data/carb/oie_readers/__init__.py +0 -0
- evaluation_data/carb/oie_readers/allennlpReader.py +86 -0
- evaluation_data/carb/oie_readers/argument.py +21 -0
- evaluation_data/carb/oie_readers/benchmarkGoldReader.py +55 -0
- evaluation_data/carb/oie_readers/clausieReader.py +90 -0
- evaluation_data/carb/oie_readers/extraction.py +443 -0
- evaluation_data/carb/oie_readers/goldReader.py +44 -0
- evaluation_data/carb/oie_readers/oieReader.py +45 -0
- evaluation_data/carb/oie_readers/ollieReader.py +22 -0
- evaluation_data/carb/oie_readers/openieFiveReader.py +38 -0
- evaluation_data/carb/oie_readers/openieFourReader.py +59 -0
- evaluation_data/carb/oie_readers/propsReader.py +44 -0
- evaluation_data/carb/oie_readers/reVerbReader.py +3 -1
- evaluation_data/carb/oie_readers/split_corpus.py +1 -1
- evaluation_data/carb/oie_readers/stanfordReader.py +1 -1
- evaluation_data/carb/oie_readers/tabReader.py +1 -1
evaluation_data/carb/oie_readers/__init__.py
ADDED
|
File without changes
|
evaluation_data/carb/oie_readers/allennlpReader.py
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from .oieReader import OieReader
|
| 2 |
+
from .extraction import Extraction
|
| 3 |
+
import math
|
| 4 |
+
import os
|
| 5 |
+
import ipdb
|
| 6 |
+
|
| 7 |
+
class AllennlpReader(OieReader):
|
| 8 |
+
|
| 9 |
+
def __init__(self, threshold=None):
|
| 10 |
+
self.name = 'Allennlp'
|
| 11 |
+
self.threshold = threshold
|
| 12 |
+
|
| 13 |
+
def read(self, fn):
|
| 14 |
+
d = {}
|
| 15 |
+
# with open(fn) as fin:
|
| 16 |
+
if os.path.exists(fn):
|
| 17 |
+
# print("reading from file")
|
| 18 |
+
fin = open(fn)
|
| 19 |
+
else:
|
| 20 |
+
# print("reading from string")
|
| 21 |
+
fin = fn.strip().split('\n')
|
| 22 |
+
|
| 23 |
+
for line in fin:
|
| 24 |
+
'''
|
| 25 |
+
data = line.strip().split('\t')
|
| 26 |
+
confidence = data[0]
|
| 27 |
+
if not all(data[2:5]):
|
| 28 |
+
continue
|
| 29 |
+
arg1, rel = [s[s.index('(') + 1:s.index(',List(')] for s in data[2:4]]
|
| 30 |
+
#args = data[4].strip().split(');')
|
| 31 |
+
#print arg2s
|
| 32 |
+
args = [s[s.index('(') + 1:s.index(',List(')] for s in data[4].strip().split(');')]
|
| 33 |
+
# if arg1 == "the younger La Flesche":
|
| 34 |
+
# print len(args)
|
| 35 |
+
text = data[5]
|
| 36 |
+
if data[1]:
|
| 37 |
+
#print arg1, rel
|
| 38 |
+
s = data[1]
|
| 39 |
+
if not (arg1 + ' ' + rel).startswith(s[s.index('(') + 1:s.index(',List(')]):
|
| 40 |
+
#print "##########Not adding context"
|
| 41 |
+
arg1 = s[s.index('(') + 1:s.index(',List(')] + ' ' + arg1
|
| 42 |
+
#print arg1 + rel, ",,,,, ", s[s.index('(') + 1:s.index(',List(')]
|
| 43 |
+
'''
|
| 44 |
+
# print(line)
|
| 45 |
+
line = line.strip().split('\t')
|
| 46 |
+
# print(line)
|
| 47 |
+
text = line[0]
|
| 48 |
+
try:
|
| 49 |
+
confidence = line[2]
|
| 50 |
+
except:
|
| 51 |
+
confidence = 0
|
| 52 |
+
# raise Exception('Unable to find confidence in line: ',line)
|
| 53 |
+
line = line[1]
|
| 54 |
+
try:
|
| 55 |
+
arg1 = line[line.index('<arg1>') + 6:line.index('</arg1>')]
|
| 56 |
+
except:
|
| 57 |
+
arg1 = ""
|
| 58 |
+
try:
|
| 59 |
+
rel = line[line.index('<rel>') + 5:line.index('</rel>')]
|
| 60 |
+
except:
|
| 61 |
+
rel = ""
|
| 62 |
+
try:
|
| 63 |
+
arg2 = line[line.index('<arg2>') + 6:line.index('</arg2>')]
|
| 64 |
+
except:
|
| 65 |
+
arg2 = ""
|
| 66 |
+
|
| 67 |
+
if(type(self.threshold) != type(None) and float(confidence) < self.threshold):
|
| 68 |
+
continue
|
| 69 |
+
|
| 70 |
+
if not ( arg1 or arg2 or rel):
|
| 71 |
+
continue
|
| 72 |
+
#confidence = 1
|
| 73 |
+
#print(arg1, rel, arg2, confidence)
|
| 74 |
+
# curExtraction = Extraction(pred = rel, head_pred_index = -1, sent = text, confidence = -1/float(confidence))
|
| 75 |
+
# curExtraction = Extraction(pred = rel, head_pred_index = -1, sent = text, confidence = math.exp(float(confidence)))
|
| 76 |
+
curExtraction = Extraction(pred = rel, head_pred_index = -1, sent = text, confidence = float(confidence))
|
| 77 |
+
curExtraction.addArg(arg1)
|
| 78 |
+
curExtraction.addArg(arg2)
|
| 79 |
+
#for arg in args:
|
| 80 |
+
# curExtraction.addArg(arg)
|
| 81 |
+
d[text] = d.get(text, []) + [curExtraction]
|
| 82 |
+
|
| 83 |
+
if os.path.exists(fn):
|
| 84 |
+
fin.close()
|
| 85 |
+
# print(d)
|
| 86 |
+
self.oie = d
|
evaluation_data/carb/oie_readers/argument.py
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import nltk
|
| 2 |
+
from operator import itemgetter
|
| 3 |
+
|
| 4 |
+
class Argument:
|
| 5 |
+
def __init__(self, arg):
|
| 6 |
+
self.words = [x for x in arg[0].strip().split(' ') if x]
|
| 7 |
+
self.posTags = map(itemgetter(1), nltk.pos_tag(self.words))
|
| 8 |
+
self.indices = arg[1]
|
| 9 |
+
self.feats = {}
|
| 10 |
+
|
| 11 |
+
def __str__(self):
|
| 12 |
+
return "({})".format('\t'.join(map(str,
|
| 13 |
+
[escape_special_chars(' '.join(self.words)),
|
| 14 |
+
str(self.indices)])))
|
| 15 |
+
|
| 16 |
+
COREF = 'coref'
|
| 17 |
+
|
| 18 |
+
## Helper functions
|
| 19 |
+
def escape_special_chars(s):
|
| 20 |
+
return s.replace('\t', '\\t')
|
| 21 |
+
|
evaluation_data/carb/oie_readers/benchmarkGoldReader.py
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
""" Usage:
|
| 2 |
+
benchmarkGoldReader --in=INPUT_FILE
|
| 3 |
+
|
| 4 |
+
Read a tab-formatted file.
|
| 5 |
+
Each line consists of:
|
| 6 |
+
sent, prob, pred, arg1, arg2, ...
|
| 7 |
+
|
| 8 |
+
"""
|
| 9 |
+
|
| 10 |
+
from oie_readers.oieReader import OieReader
|
| 11 |
+
from oie_readers.extraction import Extraction
|
| 12 |
+
from docopt import docopt
|
| 13 |
+
import logging
|
| 14 |
+
|
| 15 |
+
logging.basicConfig(level = logging.DEBUG)
|
| 16 |
+
|
| 17 |
+
class BenchmarkGoldReader(OieReader):
|
| 18 |
+
|
| 19 |
+
def __init__(self):
|
| 20 |
+
self.name = 'BenchmarkGoldReader'
|
| 21 |
+
|
| 22 |
+
def read(self, fn):
|
| 23 |
+
"""
|
| 24 |
+
Read a tabbed format line
|
| 25 |
+
Each line consists of:
|
| 26 |
+
sent, prob, pred, arg1, arg2, ...
|
| 27 |
+
"""
|
| 28 |
+
d = {}
|
| 29 |
+
ex_index = 0
|
| 30 |
+
with open(fn) as fin:
|
| 31 |
+
for line in fin:
|
| 32 |
+
if not line.strip():
|
| 33 |
+
continue
|
| 34 |
+
data = line.strip().split('\t')
|
| 35 |
+
text, rel = data[:2]
|
| 36 |
+
curExtraction = Extraction(pred = rel.strip(),
|
| 37 |
+
head_pred_index = None,
|
| 38 |
+
sent = text.strip(),
|
| 39 |
+
confidence = 1.0,
|
| 40 |
+
question_dist = "./question_distributions/dist_wh_sbj_obj1.json",
|
| 41 |
+
index = ex_index)
|
| 42 |
+
ex_index += 1
|
| 43 |
+
|
| 44 |
+
for arg in data[2:]:
|
| 45 |
+
curExtraction.addArg(arg.strip())
|
| 46 |
+
|
| 47 |
+
d[text] = d.get(text, []) + [curExtraction]
|
| 48 |
+
self.oie = d
|
| 49 |
+
|
| 50 |
+
|
| 51 |
+
if __name__ == "__main__":
|
| 52 |
+
args = docopt(__doc__)
|
| 53 |
+
input_fn = args["--in"]
|
| 54 |
+
tr = BenchmarkGoldReader()
|
| 55 |
+
tr.read(input_fn)
|
evaluation_data/carb/oie_readers/clausieReader.py
ADDED
|
@@ -0,0 +1,90 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
""" Usage:
|
| 2 |
+
<file-name> --in=INPUT_FILE --out=OUTPUT_FILE [--debug]
|
| 3 |
+
|
| 4 |
+
Convert to tabbed format
|
| 5 |
+
"""
|
| 6 |
+
# External imports
|
| 7 |
+
import logging
|
| 8 |
+
from pprint import pprint
|
| 9 |
+
from pprint import pformat
|
| 10 |
+
from docopt import docopt
|
| 11 |
+
|
| 12 |
+
# Local imports
|
| 13 |
+
from oie_readers.oieReader import OieReader
|
| 14 |
+
from oie_readers.extraction import Extraction
|
| 15 |
+
import ipdb
|
| 16 |
+
#=-----
|
| 17 |
+
|
| 18 |
+
class ClausieReader(OieReader):
|
| 19 |
+
|
| 20 |
+
def __init__(self):
|
| 21 |
+
self.name = 'ClausIE'
|
| 22 |
+
|
| 23 |
+
def read(self, fn):
|
| 24 |
+
d = {}
|
| 25 |
+
with open(fn, encoding="utf-8") as fin:
|
| 26 |
+
for line in fin:
|
| 27 |
+
data = line.strip().split('\t')
|
| 28 |
+
if len(data) == 1:
|
| 29 |
+
text = data[0]
|
| 30 |
+
elif len(data) == 5:
|
| 31 |
+
arg1, rel, arg2 = [s[1:-1] for s in data[1:4]]
|
| 32 |
+
confidence = data[4]
|
| 33 |
+
|
| 34 |
+
curExtraction = Extraction(pred = rel,
|
| 35 |
+
head_pred_index = -1,
|
| 36 |
+
sent = text,
|
| 37 |
+
confidence = float(confidence))
|
| 38 |
+
|
| 39 |
+
curExtraction.addArg(arg1)
|
| 40 |
+
curExtraction.addArg(arg2)
|
| 41 |
+
d[text] = d.get(text, []) + [curExtraction]
|
| 42 |
+
self.oie = d
|
| 43 |
+
# self.normalizeConfidence()
|
| 44 |
+
|
| 45 |
+
# # remove exxtractions below the confidence threshold
|
| 46 |
+
# if type(self.threshold) != type(None):
|
| 47 |
+
# new_d = {}
|
| 48 |
+
# for sent in self.oie:
|
| 49 |
+
# for extraction in self.oie[sent]:
|
| 50 |
+
# if extraction.confidence < self.threshold:
|
| 51 |
+
# continue
|
| 52 |
+
# else:
|
| 53 |
+
# new_d[sent] = new_d.get(sent, []) + [extraction]
|
| 54 |
+
# self.oie = new_d
|
| 55 |
+
|
| 56 |
+
|
| 57 |
+
|
| 58 |
+
def normalizeConfidence(self):
|
| 59 |
+
''' Normalize confidence to resemble probabilities '''
|
| 60 |
+
EPSILON = 1e-3
|
| 61 |
+
|
| 62 |
+
confidences = [extraction.confidence for sent in self.oie for extraction in self.oie[sent]]
|
| 63 |
+
maxConfidence = max(confidences)
|
| 64 |
+
minConfidence = min(confidences)
|
| 65 |
+
|
| 66 |
+
denom = maxConfidence - minConfidence + (2*EPSILON)
|
| 67 |
+
|
| 68 |
+
for sent, extractions in self.oie.items():
|
| 69 |
+
for extraction in extractions:
|
| 70 |
+
extraction.confidence = ( (extraction.confidence - minConfidence) + EPSILON) / denom
|
| 71 |
+
|
| 72 |
+
|
| 73 |
+
|
| 74 |
+
if __name__ == "__main__":
|
| 75 |
+
# Parse command line arguments
|
| 76 |
+
args = docopt(__doc__)
|
| 77 |
+
inp_fn = args["--in"]
|
| 78 |
+
out_fn = args["--out"]
|
| 79 |
+
debug = args["--debug"]
|
| 80 |
+
if debug:
|
| 81 |
+
logging.basicConfig(level = logging.DEBUG)
|
| 82 |
+
else:
|
| 83 |
+
logging.basicConfig(level = logging.INFO)
|
| 84 |
+
|
| 85 |
+
|
| 86 |
+
oie = ClausieReader()
|
| 87 |
+
oie.read(inp_fn)
|
| 88 |
+
oie.output_tabbed(out_fn)
|
| 89 |
+
|
| 90 |
+
logging.info("DONE")
|
evaluation_data/carb/oie_readers/extraction.py
ADDED
|
@@ -0,0 +1,443 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from oie_readers.argument import Argument
|
| 2 |
+
from operator import itemgetter
|
| 3 |
+
from collections import defaultdict
|
| 4 |
+
import nltk
|
| 5 |
+
import itertools
|
| 6 |
+
import logging
|
| 7 |
+
import numpy as np
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
class Extraction:
|
| 11 |
+
"""
|
| 12 |
+
Stores sentence, single predicate and corresponding arguments.
|
| 13 |
+
"""
|
| 14 |
+
def __init__(self, pred, head_pred_index, sent, confidence, question_dist = '', index = -1):
|
| 15 |
+
self.pred = pred
|
| 16 |
+
self.head_pred_index = head_pred_index
|
| 17 |
+
self.sent = sent
|
| 18 |
+
self.args = []
|
| 19 |
+
self.confidence = confidence
|
| 20 |
+
self.matched = []
|
| 21 |
+
self.questions = {}
|
| 22 |
+
self.indsForQuestions = defaultdict(lambda: set())
|
| 23 |
+
self.is_mwp = False
|
| 24 |
+
self.question_dist = question_dist
|
| 25 |
+
self.index = index
|
| 26 |
+
|
| 27 |
+
def distArgFromPred(self, arg):
|
| 28 |
+
assert(len(self.pred) == 2)
|
| 29 |
+
dists = []
|
| 30 |
+
for x in self.pred[1]:
|
| 31 |
+
for y in arg.indices:
|
| 32 |
+
dists.append(abs(x - y))
|
| 33 |
+
|
| 34 |
+
return min(dists)
|
| 35 |
+
|
| 36 |
+
def argsByDistFromPred(self, question):
|
| 37 |
+
return sorted(self.questions[question], key = lambda arg: self.distArgFromPred(arg))
|
| 38 |
+
|
| 39 |
+
def addArg(self, arg, question = None):
|
| 40 |
+
self.args.append(arg)
|
| 41 |
+
if question:
|
| 42 |
+
self.questions[question] = self.questions.get(question,[]) + [Argument(arg)]
|
| 43 |
+
|
| 44 |
+
def noPronounArgs(self):
|
| 45 |
+
"""
|
| 46 |
+
Returns True iff all of this extraction's arguments are not pronouns.
|
| 47 |
+
"""
|
| 48 |
+
for (a, _) in self.args:
|
| 49 |
+
tokenized_arg = nltk.word_tokenize(a)
|
| 50 |
+
if len(tokenized_arg) == 1:
|
| 51 |
+
_, pos_tag = nltk.pos_tag(tokenized_arg)[0]
|
| 52 |
+
if ('PRP' in pos_tag):
|
| 53 |
+
return False
|
| 54 |
+
return True
|
| 55 |
+
|
| 56 |
+
def isContiguous(self):
|
| 57 |
+
return all([indices for (_, indices) in self.args])
|
| 58 |
+
|
| 59 |
+
def toBinary(self):
|
| 60 |
+
''' Try to represent this extraction's arguments as binary
|
| 61 |
+
If fails, this function will return an empty list. '''
|
| 62 |
+
|
| 63 |
+
ret = [self.elementToStr(self.pred)]
|
| 64 |
+
|
| 65 |
+
if len(self.args) == 2:
|
| 66 |
+
# we're in luck
|
| 67 |
+
return ret + [self.elementToStr(arg) for arg in self.args]
|
| 68 |
+
|
| 69 |
+
return []
|
| 70 |
+
|
| 71 |
+
if not self.isContiguous():
|
| 72 |
+
# give up on non contiguous arguments (as we need indexes)
|
| 73 |
+
return []
|
| 74 |
+
|
| 75 |
+
# otherwise, try to merge based on indices
|
| 76 |
+
# TODO: you can explore other methods for doing this
|
| 77 |
+
binarized = self.binarizeByIndex()
|
| 78 |
+
|
| 79 |
+
if binarized:
|
| 80 |
+
return ret + binarized
|
| 81 |
+
|
| 82 |
+
return []
|
| 83 |
+
|
| 84 |
+
|
| 85 |
+
def elementToStr(self, elem, print_indices = True):
|
| 86 |
+
''' formats an extraction element (pred or arg) as a raw string
|
| 87 |
+
removes indices and trailing spaces '''
|
| 88 |
+
if print_indices:
|
| 89 |
+
return str(elem)
|
| 90 |
+
if isinstance(elem, str):
|
| 91 |
+
return elem
|
| 92 |
+
if isinstance(elem, tuple):
|
| 93 |
+
ret = elem[0].rstrip().lstrip()
|
| 94 |
+
else:
|
| 95 |
+
ret = ' '.join(elem.words)
|
| 96 |
+
assert ret, "empty element? {0}".format(elem)
|
| 97 |
+
return ret
|
| 98 |
+
|
| 99 |
+
def binarizeByIndex(self):
|
| 100 |
+
extraction = [self.pred] + self.args
|
| 101 |
+
markPred = [(w, ind, i == 0) for i, (w, ind) in enumerate(extraction)]
|
| 102 |
+
sortedExtraction = sorted(markPred, key = lambda ws, indices, f : indices[0])
|
| 103 |
+
s = ' '.join(['{1} {0} {1}'.format(self.elementToStr(elem), SEP) if elem[2] else self.elementToStr(elem) for elem in sortedExtraction])
|
| 104 |
+
binArgs = [a for a in s.split(SEP) if a.rstrip().lstrip()]
|
| 105 |
+
|
| 106 |
+
if len(binArgs) == 2:
|
| 107 |
+
return binArgs
|
| 108 |
+
|
| 109 |
+
# failure
|
| 110 |
+
return []
|
| 111 |
+
|
| 112 |
+
def bow(self):
|
| 113 |
+
return ' '.join([self.elementToStr(elem) for elem in [self.pred] + self.args])
|
| 114 |
+
|
| 115 |
+
def getSortedArgs(self):
|
| 116 |
+
"""
|
| 117 |
+
Sort the list of arguments.
|
| 118 |
+
If a question distribution is provided - use it,
|
| 119 |
+
otherwise, default to the order of appearance in the sentence.
|
| 120 |
+
"""
|
| 121 |
+
if self.question_dist:
|
| 122 |
+
# There's a question distribtuion - use it
|
| 123 |
+
return self.sort_args_by_distribution()
|
| 124 |
+
ls = []
|
| 125 |
+
for q, args in self.questions.iteritems():
|
| 126 |
+
if (len(args) != 1):
|
| 127 |
+
logging.debug("Not one argument: {}".format(args))
|
| 128 |
+
continue
|
| 129 |
+
arg = args[0]
|
| 130 |
+
indices = list(self.indsForQuestions[q].union(arg.indices))
|
| 131 |
+
if not indices:
|
| 132 |
+
logging.debug("Empty indexes for arg {} -- backing to zero".format(arg))
|
| 133 |
+
indices = [0]
|
| 134 |
+
ls.append(((arg, q), indices))
|
| 135 |
+
return [a for a, _ in sorted(ls,
|
| 136 |
+
key = lambda _, indices: min(indices))]
|
| 137 |
+
|
| 138 |
+
def question_prob_for_loc(self, question, loc):
|
| 139 |
+
"""
|
| 140 |
+
Returns the probability of the given question leading to argument
|
| 141 |
+
appearing in the given location in the output slot.
|
| 142 |
+
"""
|
| 143 |
+
gen_question = generalize_question(question)
|
| 144 |
+
q_dist = self.question_dist[gen_question]
|
| 145 |
+
logging.debug("distribution of {}: {}".format(gen_question,
|
| 146 |
+
q_dist))
|
| 147 |
+
|
| 148 |
+
return float(q_dist.get(loc, 0)) / \
|
| 149 |
+
sum(q_dist.values())
|
| 150 |
+
|
| 151 |
+
def sort_args_by_distribution(self):
|
| 152 |
+
"""
|
| 153 |
+
Use this instance's question distribution (this func assumes it exists)
|
| 154 |
+
in determining the positioning of the arguments.
|
| 155 |
+
Greedy algorithm:
|
| 156 |
+
0. Decide on which argument will serve as the ``subject'' (first slot) of this extraction
|
| 157 |
+
0.1 Based on the most probable one for this spot
|
| 158 |
+
(special care is given to select the highly-influential subject position)
|
| 159 |
+
1. For all other arguments, sort arguments by the prevalance of their questions
|
| 160 |
+
2. For each argument:
|
| 161 |
+
2.1 Assign to it the most probable slot still available
|
| 162 |
+
2.2 If non such exist (fallback) - default to put it in the last location
|
| 163 |
+
"""
|
| 164 |
+
INF_LOC = 100 # Used as an impractical last argument
|
| 165 |
+
|
| 166 |
+
# Store arguments by slot
|
| 167 |
+
ret = {INF_LOC: []}
|
| 168 |
+
logging.debug("sorting: {}".format(self.questions))
|
| 169 |
+
|
| 170 |
+
# Find the most suitable arguemnt for the subject location
|
| 171 |
+
logging.debug("probs for subject: {}".format([(q, self.question_prob_for_loc(q, 0))
|
| 172 |
+
for (q, _) in self.questions.iteritems()]))
|
| 173 |
+
|
| 174 |
+
subj_question, subj_args = max(self.questions.iteritems(),
|
| 175 |
+
key = lambda q, _: self.question_prob_for_loc(q, 0))
|
| 176 |
+
|
| 177 |
+
ret[0] = [(subj_args[0], subj_question)]
|
| 178 |
+
|
| 179 |
+
# Find the rest
|
| 180 |
+
for (question, args) in sorted([(q, a)
|
| 181 |
+
for (q, a) in self.questions.iteritems() if (q not in [subj_question])],
|
| 182 |
+
key = lambda q, _: \
|
| 183 |
+
sum(self.question_dist[generalize_question(q)].values()),
|
| 184 |
+
reverse = True):
|
| 185 |
+
gen_question = generalize_question(question)
|
| 186 |
+
arg = args[0]
|
| 187 |
+
assigned_flag = False
|
| 188 |
+
for (loc, count) in sorted(self.question_dist[gen_question].iteritems(),
|
| 189 |
+
key = lambda _ , c: c,
|
| 190 |
+
reverse = True):
|
| 191 |
+
if loc not in ret:
|
| 192 |
+
# Found an empty slot for this item
|
| 193 |
+
# Place it there and break out
|
| 194 |
+
ret[loc] = [(arg, question)]
|
| 195 |
+
assigned_flag = True
|
| 196 |
+
break
|
| 197 |
+
|
| 198 |
+
if not assigned_flag:
|
| 199 |
+
# Add this argument to the non-assigned (hopefully doesn't happen much)
|
| 200 |
+
logging.debug("Couldn't find an open assignment for {}".format((arg, gen_question)))
|
| 201 |
+
ret[INF_LOC].append((arg, question))
|
| 202 |
+
|
| 203 |
+
logging.debug("Linearizing arg list: {}".format(ret))
|
| 204 |
+
|
| 205 |
+
# Finished iterating - consolidate and return a list of arguments
|
| 206 |
+
return [arg
|
| 207 |
+
for (_, arg_ls) in sorted(ret.iteritems(),
|
| 208 |
+
key = lambda k, v: int(k))
|
| 209 |
+
for arg in arg_ls]
|
| 210 |
+
|
| 211 |
+
|
| 212 |
+
def __str__(self):
|
| 213 |
+
pred_str = self.elementToStr(self.pred)
|
| 214 |
+
return '{}\t{}\t{}'.format(self.get_base_verb(pred_str),
|
| 215 |
+
self.compute_global_pred(pred_str,
|
| 216 |
+
self.questions.keys()),
|
| 217 |
+
'\t'.join([escape_special_chars(self.augment_arg_with_question(self.elementToStr(arg),
|
| 218 |
+
question))
|
| 219 |
+
for arg, question in self.getSortedArgs()]))
|
| 220 |
+
|
| 221 |
+
def get_base_verb(self, surface_pred):
|
| 222 |
+
"""
|
| 223 |
+
Given the surface pred, return the original annotated verb
|
| 224 |
+
"""
|
| 225 |
+
# Assumes that at this point the verb is always the last word
|
| 226 |
+
# in the surface predicate
|
| 227 |
+
return surface_pred.split(' ')[-1]
|
| 228 |
+
|
| 229 |
+
|
| 230 |
+
def compute_global_pred(self, surface_pred, questions):
|
| 231 |
+
"""
|
| 232 |
+
Given the surface pred and all instansiations of questions,
|
| 233 |
+
make global coherence decisions regarding the final form of the predicate
|
| 234 |
+
This should hopefully take care of multi word predicates and correct inflections
|
| 235 |
+
"""
|
| 236 |
+
from operator import itemgetter
|
| 237 |
+
split_surface = surface_pred.split(' ')
|
| 238 |
+
|
| 239 |
+
if len(split_surface) > 1:
|
| 240 |
+
# This predicate has a modal preceding the base verb
|
| 241 |
+
verb = split_surface[-1]
|
| 242 |
+
ret = split_surface[:-1] # get all of the elements in the modal
|
| 243 |
+
else:
|
| 244 |
+
verb = split_surface[0]
|
| 245 |
+
ret = []
|
| 246 |
+
|
| 247 |
+
split_questions = map(lambda question: question.split(' '),
|
| 248 |
+
questions)
|
| 249 |
+
|
| 250 |
+
preds = map(normalize_element,
|
| 251 |
+
map(itemgetter(QUESTION_TRG_INDEX),
|
| 252 |
+
split_questions))
|
| 253 |
+
if len(set(preds)) > 1:
|
| 254 |
+
# This predicate is appears in multiple ways, let's stick to the base form
|
| 255 |
+
ret.append(verb)
|
| 256 |
+
|
| 257 |
+
if len(set(preds)) == 1:
|
| 258 |
+
# Change the predciate to the inflected form
|
| 259 |
+
# if there's exactly one way in which the predicate is conveyed
|
| 260 |
+
ret.append(preds[0])
|
| 261 |
+
|
| 262 |
+
pps = map(normalize_element,
|
| 263 |
+
map(itemgetter(QUESTION_PP_INDEX),
|
| 264 |
+
split_questions))
|
| 265 |
+
|
| 266 |
+
obj2s = map(normalize_element,
|
| 267 |
+
map(itemgetter(QUESTION_OBJ2_INDEX),
|
| 268 |
+
split_questions))
|
| 269 |
+
|
| 270 |
+
if (len(set(pps)) == 1):
|
| 271 |
+
# If all questions for the predicate include the same pp attachemnt -
|
| 272 |
+
# assume it's a multiword predicate
|
| 273 |
+
self.is_mwp = True # Signal to arguments that they shouldn't take the preposition
|
| 274 |
+
ret.append(pps[0])
|
| 275 |
+
|
| 276 |
+
# Concat all elements in the predicate and return
|
| 277 |
+
return " ".join(ret).strip()
|
| 278 |
+
|
| 279 |
+
|
| 280 |
+
def augment_arg_with_question(self, arg, question):
|
| 281 |
+
"""
|
| 282 |
+
Decide what elements from the question to incorporate in the given
|
| 283 |
+
corresponding argument
|
| 284 |
+
"""
|
| 285 |
+
# Parse question
|
| 286 |
+
wh, aux, sbj, trg, obj1, pp, obj2 = map(normalize_element,
|
| 287 |
+
question.split(' ')[:-1]) # Last split is the question mark
|
| 288 |
+
|
| 289 |
+
# Place preposition in argument
|
| 290 |
+
# This is safer when dealing with n-ary arguments, as it's directly attaches to the
|
| 291 |
+
# appropriate argument
|
| 292 |
+
if (not self.is_mwp) and pp and (not obj2):
|
| 293 |
+
if not(arg.startswith("{} ".format(pp))):
|
| 294 |
+
# Avoid repeating the preporition in cases where both question and answer contain it
|
| 295 |
+
return " ".join([pp,
|
| 296 |
+
arg])
|
| 297 |
+
|
| 298 |
+
# Normal cases
|
| 299 |
+
return arg
|
| 300 |
+
|
| 301 |
+
def clusterScore(self, cluster):
|
| 302 |
+
"""
|
| 303 |
+
Calculate cluster density score as the mean distance of the maximum distance of each slot.
|
| 304 |
+
Lower score represents a denser cluster.
|
| 305 |
+
"""
|
| 306 |
+
logging.debug("*-*-*- Cluster: {}".format(cluster))
|
| 307 |
+
|
| 308 |
+
# Find global centroid
|
| 309 |
+
arr = np.array([x for ls in cluster for x in ls])
|
| 310 |
+
centroid = np.sum(arr)/arr.shape[0]
|
| 311 |
+
logging.debug("Centroid: {}".format(centroid))
|
| 312 |
+
|
| 313 |
+
# Calculate mean over all maxmimum points
|
| 314 |
+
return np.average([max([abs(x - centroid) for x in ls]) for ls in cluster])
|
| 315 |
+
|
| 316 |
+
def resolveAmbiguity(self):
|
| 317 |
+
"""
|
| 318 |
+
Heursitic to map the elments (argument and predicates) of this extraction
|
| 319 |
+
back to the indices of the sentence.
|
| 320 |
+
"""
|
| 321 |
+
## TODO: This removes arguments for which there was no consecutive span found
|
| 322 |
+
## Part of these are non-consecutive arguments,
|
| 323 |
+
## but other could be a bug in recognizing some punctuation marks
|
| 324 |
+
|
| 325 |
+
elements = [self.pred] \
|
| 326 |
+
+ [(s, indices)
|
| 327 |
+
for (s, indices)
|
| 328 |
+
in self.args
|
| 329 |
+
if indices]
|
| 330 |
+
logging.debug("Resolving ambiguity in: {}".format(elements))
|
| 331 |
+
|
| 332 |
+
# Collect all possible combinations of arguments and predicate indices
|
| 333 |
+
# (hopefully it's not too much)
|
| 334 |
+
all_combinations = list(itertools.product(*map(itemgetter(1), elements)))
|
| 335 |
+
logging.debug("Number of combinations: {}".format(len(all_combinations)))
|
| 336 |
+
|
| 337 |
+
# Choose the ones with best clustering and unfold them
|
| 338 |
+
resolved_elements = zip(map(itemgetter(0), elements),
|
| 339 |
+
min(all_combinations,
|
| 340 |
+
key = lambda cluster: self.clusterScore(cluster)))
|
| 341 |
+
logging.debug("Resolved elements = {}".format(resolved_elements))
|
| 342 |
+
|
| 343 |
+
self.pred = resolved_elements[0]
|
| 344 |
+
self.args = resolved_elements[1:]
|
| 345 |
+
|
| 346 |
+
def conll(self, external_feats = {}):
|
| 347 |
+
"""
|
| 348 |
+
Return a CoNLL string representation of this extraction
|
| 349 |
+
"""
|
| 350 |
+
return '\n'.join(["\t".join(map(str,
|
| 351 |
+
[i, w] + \
|
| 352 |
+
list(self.pred) + \
|
| 353 |
+
[self.head_pred_index] + \
|
| 354 |
+
external_feats + \
|
| 355 |
+
[self.get_label(i)]))
|
| 356 |
+
for (i, w)
|
| 357 |
+
in enumerate(self.sent.split(" "))]) + '\n'
|
| 358 |
+
|
| 359 |
+
def get_label(self, index):
|
| 360 |
+
"""
|
| 361 |
+
Given an index of a word in the sentence -- returns the appropriate BIO conll label
|
| 362 |
+
Assumes that ambiguation was already resolved.
|
| 363 |
+
"""
|
| 364 |
+
# Get the element(s) in which this index appears
|
| 365 |
+
ent = [(elem_ind, elem)
|
| 366 |
+
for (elem_ind, elem)
|
| 367 |
+
in enumerate(map(itemgetter(1),
|
| 368 |
+
[self.pred] + self.args))
|
| 369 |
+
if index in elem]
|
| 370 |
+
|
| 371 |
+
if not ent:
|
| 372 |
+
# index doesnt appear in any element
|
| 373 |
+
return "O"
|
| 374 |
+
|
| 375 |
+
if len(ent) > 1:
|
| 376 |
+
# The same word appears in two different answers
|
| 377 |
+
# In this case we choose the first one as label
|
| 378 |
+
logging.warn("Index {} appears in one than more element: {}".\
|
| 379 |
+
format(index,
|
| 380 |
+
"\t".join(map(str,
|
| 381 |
+
[ent,
|
| 382 |
+
self.sent,
|
| 383 |
+
self.pred,
|
| 384 |
+
self.args]))))
|
| 385 |
+
|
| 386 |
+
## Some indices appear in more than one argument (ones where the above message appears)
|
| 387 |
+
## From empricial observation, these seem to mostly consist of different levels of granularity:
|
| 388 |
+
## what had _ been taken _ _ _ ? loan commitments topping $ 3 billion
|
| 389 |
+
## how much had _ been taken _ _ _ ? topping $ 3 billion
|
| 390 |
+
## In these cases we heuristically choose the shorter answer span, hopefully creating minimal spans
|
| 391 |
+
## E.g., in this example two arguemnts are created: (loan commitments, topping $ 3 billion)
|
| 392 |
+
|
| 393 |
+
elem_ind, elem = min(ent, key = lambda _, ls: len(ls))
|
| 394 |
+
|
| 395 |
+
# Distinguish between predicate and arguments
|
| 396 |
+
prefix = "P" if elem_ind == 0 else "A{}".format(elem_ind - 1)
|
| 397 |
+
|
| 398 |
+
# Distinguish between Beginning and Inside labels
|
| 399 |
+
suffix = "B" if index == elem[0] else "I"
|
| 400 |
+
|
| 401 |
+
return "{}-{}".format(prefix, suffix)
|
| 402 |
+
|
| 403 |
+
def __str__(self):
|
| 404 |
+
return '{0}\t{1}'.format(self.elementToStr(self.pred,
|
| 405 |
+
print_indices = True),
|
| 406 |
+
'\t'.join([self.elementToStr(arg)
|
| 407 |
+
for arg
|
| 408 |
+
in self.args]))
|
| 409 |
+
|
| 410 |
+
# Flatten a list of lists
|
| 411 |
+
flatten = lambda l: [item for sublist in l for item in sublist]
|
| 412 |
+
|
| 413 |
+
|
| 414 |
+
def normalize_element(elem):
|
| 415 |
+
"""
|
| 416 |
+
Return a surface form of the given question element.
|
| 417 |
+
the output should be properly able to precede a predicate (or blank otherwise)
|
| 418 |
+
"""
|
| 419 |
+
return elem.replace("_", " ") \
|
| 420 |
+
if (elem != "_")\
|
| 421 |
+
else ""
|
| 422 |
+
|
| 423 |
+
## Helper functions
|
| 424 |
+
def escape_special_chars(s):
|
| 425 |
+
return s.replace('\t', '\\t')
|
| 426 |
+
|
| 427 |
+
|
| 428 |
+
def generalize_question(question):
|
| 429 |
+
"""
|
| 430 |
+
Given a question in the context of the sentence and the predicate index within
|
| 431 |
+
the question - return a generalized version which extracts only order-imposing features
|
| 432 |
+
"""
|
| 433 |
+
import nltk # Using nltk since couldn't get spaCy to agree on the tokenization
|
| 434 |
+
wh, aux, sbj, trg, obj1, pp, obj2 = question.split(' ')[:-1] # Last split is the question mark
|
| 435 |
+
return ' '.join([wh, sbj, obj1])
|
| 436 |
+
|
| 437 |
+
|
| 438 |
+
|
| 439 |
+
## CONSTANTS
|
| 440 |
+
SEP = ';;;'
|
| 441 |
+
QUESTION_TRG_INDEX = 3 # index of the predicate within the question
|
| 442 |
+
QUESTION_PP_INDEX = 5
|
| 443 |
+
QUESTION_OBJ2_INDEX = 6
|
evaluation_data/carb/oie_readers/goldReader.py
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from oie_readers.oieReader import OieReader
|
| 2 |
+
from oie_readers.extraction import Extraction
|
| 3 |
+
from _collections import defaultdict
|
| 4 |
+
import ipdb
|
| 5 |
+
|
| 6 |
+
class GoldReader(OieReader):
|
| 7 |
+
|
| 8 |
+
# Path relative to repo root folder
|
| 9 |
+
default_filename = './oie_corpus/all.oie'
|
| 10 |
+
|
| 11 |
+
def __init__(self):
|
| 12 |
+
self.name = 'Gold'
|
| 13 |
+
|
| 14 |
+
def read(self, fn):
|
| 15 |
+
d = defaultdict(lambda: [])
|
| 16 |
+
with open(fn) as fin:
|
| 17 |
+
for line_ind, line in enumerate(fin):
|
| 18 |
+
# print line
|
| 19 |
+
data = line.strip().split('\t')
|
| 20 |
+
text, rel = data[:2]
|
| 21 |
+
args = data[2:]
|
| 22 |
+
confidence = 1
|
| 23 |
+
|
| 24 |
+
curExtraction = Extraction(pred = rel.strip(),
|
| 25 |
+
head_pred_index = None,
|
| 26 |
+
sent = text.strip(),
|
| 27 |
+
confidence = float(confidence),
|
| 28 |
+
index = line_ind)
|
| 29 |
+
for arg in args:
|
| 30 |
+
if "C: " in arg:
|
| 31 |
+
continue
|
| 32 |
+
curExtraction.addArg(arg.strip())
|
| 33 |
+
|
| 34 |
+
d[text.strip()].append(curExtraction)
|
| 35 |
+
self.oie = d
|
| 36 |
+
|
| 37 |
+
|
| 38 |
+
if __name__ == '__main__' :
|
| 39 |
+
g = GoldReader()
|
| 40 |
+
g.read('../oie_corpus/all.oie', includeNominal = False)
|
| 41 |
+
d = g.oie
|
| 42 |
+
e = d.items()[0]
|
| 43 |
+
print(e[1][0].bow())
|
| 44 |
+
print(g.count())
|
evaluation_data/carb/oie_readers/oieReader.py
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
class OieReader:
|
| 2 |
+
|
| 3 |
+
def read(self, fn, includeNominal):
|
| 4 |
+
''' should set oie as a class member
|
| 5 |
+
as a dictionary of extractions by sentence'''
|
| 6 |
+
raise Exception("Don't run me")
|
| 7 |
+
|
| 8 |
+
def count(self):
|
| 9 |
+
''' number of extractions '''
|
| 10 |
+
return sum([len(extractions) for _, extractions in self.oie.items()])
|
| 11 |
+
|
| 12 |
+
def split_to_corpus(self, corpus_fn, out_fn):
|
| 13 |
+
"""
|
| 14 |
+
Given a corpus file name, containing a list of sentences
|
| 15 |
+
print only the extractions pertaining to it to out_fn in a tab separated format:
|
| 16 |
+
sent, prob, pred, arg1, arg2, ...
|
| 17 |
+
"""
|
| 18 |
+
raw_sents = [line.strip() for line in open(corpus_fn)]
|
| 19 |
+
with open(out_fn, 'w') as fout:
|
| 20 |
+
for line in self.get_tabbed().split('\n'):
|
| 21 |
+
data = line.split('\t')
|
| 22 |
+
sent = data[0]
|
| 23 |
+
if sent in raw_sents:
|
| 24 |
+
fout.write(line + '\n')
|
| 25 |
+
|
| 26 |
+
def output_tabbed(self, out_fn):
|
| 27 |
+
"""
|
| 28 |
+
Write a tabbed represenation of this corpus.
|
| 29 |
+
"""
|
| 30 |
+
with open(out_fn, 'w') as fout:
|
| 31 |
+
fout.write(self.get_tabbed())
|
| 32 |
+
|
| 33 |
+
def get_tabbed(self):
|
| 34 |
+
"""
|
| 35 |
+
Get a tabbed format representation of this corpus (assumes that input was
|
| 36 |
+
already read).
|
| 37 |
+
"""
|
| 38 |
+
return "\n".join(['\t'.join(map(str,
|
| 39 |
+
[ex.sent,
|
| 40 |
+
ex.confidence,
|
| 41 |
+
ex.pred,
|
| 42 |
+
'\t'.join(ex.args)]))
|
| 43 |
+
for (sent, exs) in self.oie.iteritems()
|
| 44 |
+
for ex in exs])
|
| 45 |
+
|
evaluation_data/carb/oie_readers/ollieReader.py
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from oie_readers.oieReader import OieReader
|
| 2 |
+
from oie_readers.extraction import Extraction
|
| 3 |
+
|
| 4 |
+
class OllieReader(OieReader):
|
| 5 |
+
|
| 6 |
+
def __init__(self):
|
| 7 |
+
self.name = 'OLLIE'
|
| 8 |
+
|
| 9 |
+
def read(self, fn):
|
| 10 |
+
d = {}
|
| 11 |
+
with open(fn) as fin:
|
| 12 |
+
fin.readline() #remove header
|
| 13 |
+
for line in fin:
|
| 14 |
+
data = line.strip().split('\t')
|
| 15 |
+
confidence, arg1, rel, arg2, enabler, attribution, text = data[:7]
|
| 16 |
+
curExtraction = Extraction(pred = rel, head_pred_index = -1, sent = text, confidence = float(confidence))
|
| 17 |
+
curExtraction.addArg(arg1)
|
| 18 |
+
curExtraction.addArg(arg2)
|
| 19 |
+
d[text] = d.get(text, []) + [curExtraction]
|
| 20 |
+
self.oie = d
|
| 21 |
+
|
| 22 |
+
|
evaluation_data/carb/oie_readers/openieFiveReader.py
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from oie_readers.oieReader import OieReader
|
| 2 |
+
from oie_readers.extraction import Extraction
|
| 3 |
+
|
| 4 |
+
class OpenieFiveReader(OieReader):
|
| 5 |
+
|
| 6 |
+
def __init__(self):
|
| 7 |
+
self.name = 'OpenIE-5'
|
| 8 |
+
|
| 9 |
+
def read(self, fn):
|
| 10 |
+
d = {}
|
| 11 |
+
with open(fn) as fin:
|
| 12 |
+
for line in fin:
|
| 13 |
+
data = line.strip().split('\t')
|
| 14 |
+
confidence = data[0]
|
| 15 |
+
|
| 16 |
+
if not all(data[2:5]):
|
| 17 |
+
continue
|
| 18 |
+
arg1, rel = [s[s.index('(') + 1:s.index(',List(')] for s in data[2:4]]
|
| 19 |
+
#args = data[4].strip().split(');')
|
| 20 |
+
#print arg2s
|
| 21 |
+
args = [s[s.index('(') + 1:s.index(',List(')] for s in data[4].strip().split(');')]
|
| 22 |
+
# if arg1 == "the younger La Flesche":
|
| 23 |
+
# print len(args)
|
| 24 |
+
text = data[5]
|
| 25 |
+
if data[1]:
|
| 26 |
+
#print arg1, rel
|
| 27 |
+
s = data[1]
|
| 28 |
+
if not (arg1 + ' ' + rel).startswith(s[s.index('(') + 1:s.index(',List(')]):
|
| 29 |
+
#print "##########Not adding context"
|
| 30 |
+
arg1 = s[s.index('(') + 1:s.index(',List(')] + ' ' + arg1
|
| 31 |
+
#print arg1 + rel, ",,,,, ", s[s.index('(') + 1:s.index(',List(')]
|
| 32 |
+
#curExtraction = Extraction(pred = rel, sent = text, confidence = float(confidence))
|
| 33 |
+
curExtraction = Extraction(pred = rel, head_pred_index = -1, sent = text, confidence = float(confidence))
|
| 34 |
+
curExtraction.addArg(arg1)
|
| 35 |
+
for arg in args:
|
| 36 |
+
curExtraction.addArg(arg)
|
| 37 |
+
d[text] = d.get(text, []) + [curExtraction]
|
| 38 |
+
self.oie = d
|
evaluation_data/carb/oie_readers/openieFourReader.py
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
""" Usage:
|
| 2 |
+
<file-name> --in=INPUT_FILE --out=OUTPUT_FILE [--debug]
|
| 3 |
+
|
| 4 |
+
Convert to tabbed format
|
| 5 |
+
"""
|
| 6 |
+
# External imports
|
| 7 |
+
import logging
|
| 8 |
+
from pprint import pprint
|
| 9 |
+
from pprint import pformat
|
| 10 |
+
from docopt import docopt
|
| 11 |
+
|
| 12 |
+
# Local imports
|
| 13 |
+
from oie_readers.oieReader import OieReader
|
| 14 |
+
from oie_readers.extraction import Extraction
|
| 15 |
+
import ipdb
|
| 16 |
+
|
| 17 |
+
#=-----
|
| 18 |
+
|
| 19 |
+
class OpenieFourReader(OieReader):
|
| 20 |
+
|
| 21 |
+
def __init__(self):
|
| 22 |
+
self.name = 'OpenIE-4'
|
| 23 |
+
|
| 24 |
+
def read(self, fn):
|
| 25 |
+
d = {}
|
| 26 |
+
with open(fn) as fin:
|
| 27 |
+
for line in fin:
|
| 28 |
+
data = line.strip().split('\t')
|
| 29 |
+
confidence = data[0]
|
| 30 |
+
if not all(data[2:5]):
|
| 31 |
+
logging.debug("Skipped line: {}".format(line))
|
| 32 |
+
continue
|
| 33 |
+
arg1, rel, arg2 = [s[s.index('(') + 1:s.index(',List(')] for s in data[2:5]]
|
| 34 |
+
text = data[5]
|
| 35 |
+
curExtraction = Extraction(pred = rel, head_pred_index = -1, sent = text, confidence = float(confidence))
|
| 36 |
+
curExtraction.addArg(arg1)
|
| 37 |
+
curExtraction.addArg(arg2)
|
| 38 |
+
d[text] = d.get(text, []) + [curExtraction]
|
| 39 |
+
self.oie = d
|
| 40 |
+
|
| 41 |
+
|
| 42 |
+
|
| 43 |
+
if __name__ == "__main__":
|
| 44 |
+
# Parse command line arguments
|
| 45 |
+
args = docopt(__doc__)
|
| 46 |
+
inp_fn = args["--in"]
|
| 47 |
+
out_fn = args["--out"]
|
| 48 |
+
debug = args["--debug"]
|
| 49 |
+
if debug:
|
| 50 |
+
logging.basicConfig(level = logging.DEBUG)
|
| 51 |
+
else:
|
| 52 |
+
logging.basicConfig(level = logging.INFO)
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
oie = OpenieFourReader()
|
| 56 |
+
oie.read(inp_fn)
|
| 57 |
+
oie.output_tabbed(out_fn)
|
| 58 |
+
|
| 59 |
+
logging.info("DONE")
|
evaluation_data/carb/oie_readers/propsReader.py
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from oie_readers.oieReader import OieReader
|
| 2 |
+
from oie_readers.extraction import Extraction
|
| 3 |
+
|
| 4 |
+
|
| 5 |
+
class PropSReader(OieReader):
|
| 6 |
+
|
| 7 |
+
def __init__(self):
|
| 8 |
+
self.name = 'PropS'
|
| 9 |
+
|
| 10 |
+
def read(self, fn):
|
| 11 |
+
d = {}
|
| 12 |
+
with open(fn) as fin:
|
| 13 |
+
for line in fin:
|
| 14 |
+
if not line.strip():
|
| 15 |
+
continue
|
| 16 |
+
data = line.strip().split('\t')
|
| 17 |
+
confidence, text, rel = data[:3]
|
| 18 |
+
curExtraction = Extraction(pred = rel, sent = text, confidence = float(confidence), head_pred_index=-1)
|
| 19 |
+
|
| 20 |
+
for arg in data[4::2]:
|
| 21 |
+
curExtraction.addArg(arg)
|
| 22 |
+
|
| 23 |
+
d[text] = d.get(text, []) + [curExtraction]
|
| 24 |
+
self.oie = d
|
| 25 |
+
# self.normalizeConfidence()
|
| 26 |
+
|
| 27 |
+
|
| 28 |
+
def normalizeConfidence(self):
|
| 29 |
+
''' Normalize confidence to resemble probabilities '''
|
| 30 |
+
EPSILON = 1e-3
|
| 31 |
+
|
| 32 |
+
self.confidences = [extraction.confidence for sent in self.oie for extraction in self.oie[sent]]
|
| 33 |
+
maxConfidence = max(self.confidences)
|
| 34 |
+
minConfidence = min(self.confidences)
|
| 35 |
+
|
| 36 |
+
denom = maxConfidence - minConfidence + (2*EPSILON)
|
| 37 |
+
|
| 38 |
+
for sent, extractions in self.oie.items():
|
| 39 |
+
for extraction in extractions:
|
| 40 |
+
extraction.confidence = ( (extraction.confidence - minConfidence) + EPSILON) / denom
|
| 41 |
+
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
|
evaluation_data/carb/oie_readers/reVerbReader.py
CHANGED
|
@@ -24,4 +24,6 @@ class ReVerbReader(OieReader):
|
|
| 24 |
|
| 25 |
# ReVerb requires a different files from which to get the input sentences
|
| 26 |
# Relative to repo root folder
|
| 27 |
-
RAW_SENTS_FILE = './raw_sentences/all.txt'
|
|
|
|
|
|
|
|
|
| 24 |
|
| 25 |
# ReVerb requires a different files from which to get the input sentences
|
| 26 |
# Relative to repo root folder
|
| 27 |
+
RAW_SENTS_FILE = './raw_sentences/all.txt'
|
| 28 |
+
|
| 29 |
+
|
evaluation_data/carb/oie_readers/split_corpus.py
CHANGED
|
@@ -34,4 +34,4 @@ if __name__ == "__main__":
|
|
| 34 |
reader = available_readers[args["--reader"]]()
|
| 35 |
reader.read(inp)
|
| 36 |
reader.split_to_corpus(corpus,
|
| 37 |
-
out)
|
|
|
|
| 34 |
reader = available_readers[args["--reader"]]()
|
| 35 |
reader.read(inp)
|
| 36 |
reader.split_to_corpus(corpus,
|
| 37 |
+
out)
|
evaluation_data/carb/oie_readers/stanfordReader.py
CHANGED
|
@@ -19,4 +19,4 @@ class StanfordReader(OieReader):
|
|
| 19 |
curExtraction.addArg(arg1)
|
| 20 |
curExtraction.addArg(arg2)
|
| 21 |
d[text] = d.get(text, []) + [curExtraction]
|
| 22 |
-
self.oie = d
|
|
|
|
| 19 |
curExtraction.addArg(arg1)
|
| 20 |
curExtraction.addArg(arg2)
|
| 21 |
d[text] = d.get(text, []) + [curExtraction]
|
| 22 |
+
self.oie = d
|
evaluation_data/carb/oie_readers/tabReader.py
CHANGED
|
@@ -53,4 +53,4 @@ if __name__ == "__main__":
|
|
| 53 |
args = docopt(__doc__)
|
| 54 |
input_fn = args["--in"]
|
| 55 |
tr = TabReader()
|
| 56 |
-
tr.read(input_fn)
|
|
|
|
| 53 |
args = docopt(__doc__)
|
| 54 |
input_fn = args["--in"]
|
| 55 |
tr = TabReader()
|
| 56 |
+
tr.read(input_fn)
|