dev-benchmarks-bofang

.appveyor.yml

@@ -3,9 +3,9 @@ environment:
     CMD_IN_ENV: "cmd /E:ON /V:ON /C .\\appveyor\\run_with_env.cmd"
 
   matrix:
-    - PYTHON: "C:\\Python35"
-      PYTHON_VERSION: "3.5.0"
-      PYTHON_ARCH: "32"
+#    - PYTHON: "C:\\Python35"
+#      PYTHON_VERSION: "3.5.0"
+#      PYTHON_ARCH: "32"
 
     - PYTHON: "C:\\Python35-x64"
       PYTHON_VERSION: "3.5.0"

requirements.txt

@@ -13,4 +13,6 @@ system-query
 tables
 traitlets
 tqdm
+keras
+tensorflow
 requests

tests/data/benchmarks/relation_extraction/test.txt

@@ -0,0 +1,32 @@
+Component-Whole(e2,e1)	12	15	The system as described above has its greatest application in an arrayed configuration of antenna elements .
+Other	1	9	The child was carefully wrapped and bound into the cradle by means of a cord .
+Instrument-Agency(e2,e1)	1	7	The author of a keygen uses a disassembler to look at the raw assembly code .
+Other	2	6	A misty ridge uprises from the surge .
+Member-Collection(e1,e2)	1	2	The student association is the voice of the undergraduate student population of the State University of New York at Buffalo .
+Other	4	10	This is the sprawling complex that is Peru 's largest producer of silver .
+Cause-Effect(e2,e1)	7	19	The current view is that the chronic inflammation in the distal part of the stomach caused by Helicobacter pylori infection results in an increased acid production from the non-infected upper corpus region of the stomach .
+Entity-Destination(e1,e2)	0	6	People have been moving back into downtown .
+Content-Container(e1,e2)	1	6	The lawsonite was contained in a platinum crucible and the counter-weight was a plastic crucible with metal pieces .
+Entity-Destination(e1,e2)	12	20	The solute was placed inside a beaker and 5 mL of the solvent was pipetted into a 25 mL glass flask for each trial .
+Member-Collection(e1,e2)	2	6	The fifty essays collected in this volume testify to most of the prominent themes from Professor Quispel 's scholarly career .
+Other	1	5	Their composer has sunk into oblivion .
+Message-Topic(e1,e2)	6	9	The Pulitzer Committee issues an official citation explaining the reasons for the award .
+Cause-Effect(e2,e1)	1	8	The burst has been caused by water hammer pressure .
+Instrument-Agency(e2,e1)	2	6	Even commercial networks have moved into high-definition broadcast .
+Message-Topic(e1,e2)	4	10	It was a friendly call to remind them about the bill and make sure they have a copy of the invoice .
+Instrument-Agency(e2,e1)	1	8	Texas-born virtuoso finds harmony , sophistication in Appalachian instrument .
+Product-Producer(e2,e1)	1	14	The factory 's products have included flower pots , Finnish rooster-whistles , pans , trays , tea pots , ash trays and air moisturisers .
+Component-Whole(e2,e1)	7	8	The girl showed a photo of apple tree blossom on a fruit tree in the Central Valley .
+Member-Collection(e2,e1)	20	23	They tried an assault of their own an hour later , with two columns of sixteen tanks backed by a battalion of Panzer grenadiers .
+Entity-Origin(e1,e2)	1	18	Their knowledge of the power and rank symbols of the Continental empires was gained from the numerous Germanic recruits in the Roman army , and from the Roman practice of enfeoffing various Germanic warrior groups with land in the imperial provinces .
+Member-Collection(e2,e1)	4	9	She soon had a stable of her own rescued hounds .
+Cause-Effect(e1,e2)	1	14	The singer , who performed three of the nominated songs , also caused a commotion on the red carpet .
+Other	5	11	His intellectually engaging books and essays remain pertinent to illuminating contemporary history .
+Member-Collection(e2,e1)	7	10	Poor hygiene controls , reports of a brace of gamey grouse and what looked like a skinned fox all amounted to a pie that was unfit for human consumption .
+Other	2	7	This sweet dress is made with a blend of cotton and silk , and the crochet flower necklace is the perfect accessory .
+Cause-Effect(e1,e2)	0	8	Suicide is one of the leading causes of death among pre-adolescents and teens , and victims of bullying are at an increased risk for committing suicide .
+Message-Topic(e1,e2)	1	7	This article gives details on 2004 in music in the United Kingdom , including the official charts from that year .
+Message-Topic(e1,e2)	12	16	We have therefore taken the initiative to convene the first international open meeting dedicated solely to rural history .
+Component-Whole(e1,e2)	1	4	The timer of the device automatically eliminates wasted `` standby power '' consumption by automatically turn off electronics plugged into the `` auto off '' outlets .
+Message-Topic(e2,e1)	5	8	Bob Parks made a similar offer in a phone call made earlier this week .
+Cause-Effect(e2,e1)	5	7	He had chest pains and headaches from mold in the bedrooms .

tests/data/benchmarks/relation_extraction/train.txt

@@ -0,0 +1,32 @@
+Component-Whole(e2,e1)	12	15	The system as described above has its greatest application in an arrayed configuration of antenna elements .
+Other	1	9	The child was carefully wrapped and bound into the cradle by means of a cord .
+Instrument-Agency(e2,e1)	1	7	The author of a keygen uses a disassembler to look at the raw assembly code .
+Other	2	6	A misty ridge uprises from the surge .
+Member-Collection(e1,e2)	1	2	The student association is the voice of the undergraduate student population of the State University of New York at Buffalo .
+Other	4	10	This is the sprawling complex that is Peru 's largest producer of silver .
+Cause-Effect(e2,e1)	7	19	The current view is that the chronic inflammation in the distal part of the stomach caused by Helicobacter pylori infection results in an increased acid production from the non-infected upper corpus region of the stomach .
+Entity-Destination(e1,e2)	0	6	People have been moving back into downtown .
+Content-Container(e1,e2)	1	6	The lawsonite was contained in a platinum crucible and the counter-weight was a plastic crucible with metal pieces .
+Entity-Destination(e1,e2)	12	20	The solute was placed inside a beaker and 5 mL of the solvent was pipetted into a 25 mL glass flask for each trial .
+Member-Collection(e1,e2)	2	6	The fifty essays collected in this volume testify to most of the prominent themes from Professor Quispel 's scholarly career .
+Other	1	5	Their composer has sunk into oblivion .
+Message-Topic(e1,e2)	6	9	The Pulitzer Committee issues an official citation explaining the reasons for the award .
+Cause-Effect(e2,e1)	1	8	The burst has been caused by water hammer pressure .
+Instrument-Agency(e2,e1)	2	6	Even commercial networks have moved into high-definition broadcast .
+Message-Topic(e1,e2)	4	10	It was a friendly call to remind them about the bill and make sure they have a copy of the invoice .
+Instrument-Agency(e2,e1)	1	8	Texas-born virtuoso finds harmony , sophistication in Appalachian instrument .
+Product-Producer(e2,e1)	1	14	The factory 's products have included flower pots , Finnish rooster-whistles , pans , trays , tea pots , ash trays and air moisturisers .
+Component-Whole(e2,e1)	7	8	The girl showed a photo of apple tree blossom on a fruit tree in the Central Valley .
+Member-Collection(e2,e1)	20	23	They tried an assault of their own an hour later , with two columns of sixteen tanks backed by a battalion of Panzer grenadiers .
+Entity-Origin(e1,e2)	1	18	Their knowledge of the power and rank symbols of the Continental empires was gained from the numerous Germanic recruits in the Roman army , and from the Roman practice of enfeoffing various Germanic warrior groups with land in the imperial provinces .
+Member-Collection(e2,e1)	4	9	She soon had a stable of her own rescued hounds .
+Cause-Effect(e1,e2)	1	14	The singer , who performed three of the nominated songs , also caused a commotion on the red carpet .
+Other	5	11	His intellectually engaging books and essays remain pertinent to illuminating contemporary history .
+Member-Collection(e2,e1)	7	10	Poor hygiene controls , reports of a brace of gamey grouse and what looked like a skinned fox all amounted to a pie that was unfit for human consumption .
+Other	2	7	This sweet dress is made with a blend of cotton and silk , and the crochet flower necklace is the perfect accessory .
+Cause-Effect(e1,e2)	0	8	Suicide is one of the leading causes of death among pre-adolescents and teens , and victims of bullying are at an increased risk for committing suicide .
+Message-Topic(e1,e2)	1	7	This article gives details on 2004 in music in the United Kingdom , including the official charts from that year .
+Message-Topic(e1,e2)	12	16	We have therefore taken the initiative to convene the first international open meeting dedicated solely to rural history .
+Component-Whole(e1,e2)	1	4	The timer of the device automatically eliminates wasted `` standby power '' consumption by automatically turn off electronics plugged into the `` auto off '' outlets .
+Message-Topic(e2,e1)	5	8	Bob Parks made a similar offer in a phone call made earlier this week .
+Cause-Effect(e2,e1)	5	7	He had chest pains and headaches from mold in the bedrooms .

vecto/benchmarks/language_modeling/__main__.py

@@ -24,11 +24,12 @@ def main():
                         help='use small test dataset')
     parser.add_argument("--method", default='lstm', choices=['lr', '2FFNN', 'lstm'],
                         help='name of method')
+    parser.add_argument('--normalize', dest='normalize', action='store_true')
     parser.add_argument("--path_out", default=False, help="destination folder to save results")
     args = parser.parse_args()
     embeddings = load_from_dir(args.embeddings)
     # print("embeddings", embeddings)
-    language_modeling = Language_modeling(window_size=args.window_size, method=args.method, test=args.test)
+    language_modeling = Language_modeling(normalize=args.normalize, window_size=args.window_size, method=args.method, test=args.test)
     results = language_modeling.get_result(embeddings)
     if args.path_out:
         if os.path.isdir(args.path_out) or args.path_out.endswith("/"):

vecto/benchmarks/relation_extraction/__init__.py

@@ -0,0 +1 @@
+from .relation_extraction import Relation_extraction

vecto/benchmarks/relation_extraction/__main__.py

@@ -0,0 +1,60 @@
+import argparse
+import json
+import logging
+import os
+
+from vecto.utils.data import save_json
+from vecto.benchmarks.relation_extraction import Relation_extraction
+from vecto.embeddings import load_from_dir
+
+logging.basicConfig(level=logging.DEBUG)
+
+
+def print_json(data):
+    print(json.dumps(data, ensure_ascii=False, indent=4, sort_keys=False))
+
+
+def main():
+    # config = load_config()
+    # print(config)
+    parser = argparse.ArgumentParser()
+    parser.add_argument("embeddings")
+    parser.add_argument("dataset")
+
+    parser.add_argument('--batchsize', '-b', type=int, default=64,
+                        help='Number of images in each mini-batch')
+    parser.add_argument('--epoch', '-e', type=int, default=1,
+                        help='Number of sweeps over the dataset to train')
+    parser.add_argument('--nb_filter', '-nf', type=int, default=100,
+                        help='filter number')
+    parser.add_argument('--filter_length', '-fl', type=int, default=3,
+                        help='filter length')
+    parser.add_argument('--hidden_dims', '-hd', type=int, default=100,
+                        help='D')
+    parser.add_argument('--position_dims', '-pd', type=int, default=100,
+                        help='D')
+    parser.add_argument("--path_out", default=False, help="destination folder to save results")
+    args = parser.parse_args()
+    embeddings = load_from_dir(args.embeddings)
+    # print("embeddings", embeddings)
+    # print(args.normalize)
+    relation_extraction = Relation_extraction(batchsize=args.batchsize,
+                                              epoch=args.epoch,
+                                              nb_filter=args.nb_filter,
+                                              filter_length=args.filter_length,
+                                              hidden_dims=args.hidden_dims,
+                                              position_dims=args.position_dims,)
+    results = relation_extraction.get_result(embeddings, args.dataset)
+    if args.path_out:
+        if os.path.isdir(args.path_out) or args.path_out.endswith("/"):
+            dataset = os.path.basename(os.path.normpath(args.dataset))
+            name_file_out = os.path.join(args.path_out, dataset, "results.json")
+            save_json(results, name_file_out)
+        else:
+            save_json(results, args.path_out)
+    else:
+        print_json(results)
+
+
+if __name__ == "__main__":
+    main()

vecto/benchmarks/relation_extraction/preprocess.py

@@ -0,0 +1,150 @@
+
+from __future__ import print_function
+import numpy as np
+import gzip
+import os
+import sys
+import pickle as pkl
+
+
+
+
+
+#Mapping of the labels to integers
+labelsMapping = {'Other':0,
+                 'Message-Topic(e1,e2)':1, 'Message-Topic(e2,e1)':2,
+                 'Product-Producer(e1,e2)':3, 'Product-Producer(e2,e1)':4,
+                 'Instrument-Agency(e1,e2)':5, 'Instrument-Agency(e2,e1)':6,
+                 'Entity-Destination(e1,e2)':7, 'Entity-Destination(e2,e1)':8,
+                 'Cause-Effect(e1,e2)':9, 'Cause-Effect(e2,e1)':10,
+                 'Component-Whole(e1,e2)':11, 'Component-Whole(e2,e1)':12,
+                 'Entity-Origin(e1,e2)':13, 'Entity-Origin(e2,e1)':14,
+                 'Member-Collection(e1,e2)':15, 'Member-Collection(e2,e1)':16,
+                 'Content-Container(e1,e2)':17, 'Content-Container(e2,e1)':18}
+
+
+
+
+words = {}
+maxSentenceLen = [0,0]
+
+
+distanceMapping = {'PADDING': 0, 'LowerMin': 1, 'GreaterMax': 2}
+minDistance = -30
+maxDistance = 30
+for dis in range(minDistance,maxDistance+1):
+    distanceMapping[dis] = len(distanceMapping)
+print(distanceMapping)
+
+
+def getWordIdx(token, word2Idx):
+    """Returns from the word2Idex table the word index for a given token"""
+    if token in word2Idx:
+        return word2Idx[token]
+    elif token.lower() in word2Idx:
+        return word2Idx[token.lower()]
+    return 0
+
+def createTensor(file, word2Idx, maxSentenceLen=100):
+    """Creates matrices for the events and sentence for the given file"""
+    labels = []
+    positionMatrix1 = []
+    positionMatrix2 = []
+    tokenMatrix = []
+
+    for line in open(file):
+        splits = line.strip().split('\t')
+
+        label = splits[0]
+        pos1 = splits[1]
+        pos2 = splits[2]
+        sentence = splits[3]
+        tokens = sentence.split(" ")
+
+        #print(label, pos1, pos2, sentence, tokens)
+
+
+        tokenIds = np.zeros(maxSentenceLen)
+        positionValues1 = np.zeros(maxSentenceLen)
+        positionValues2 = np.zeros(maxSentenceLen)
+
+        for idx in range(0, min(maxSentenceLen, len(tokens))):
+            tokenIds[idx] = getWordIdx(tokens[idx], word2Idx)
+
+            distance1 = idx - int(pos1)
+            distance2 = idx - int(pos2)
+            #print(distance1, distance2)
+            if distance1 in distanceMapping:
+                #print('helo')
+                positionValues1[idx] = distanceMapping[distance1]
+            elif distance1 <= minDistance:
+                positionValues1[idx] = distanceMapping['LowerMin']
+            else:
+                positionValues1[idx] = distanceMapping['GreaterMax']
+
+            if distance2 in distanceMapping:
+                positionValues2[idx] = distanceMapping[distance2]
+            elif distance2 <= minDistance:
+                positionValues2[idx] = distanceMapping['LowerMin']
+            else:
+                positionValues2[idx] = distanceMapping['GreaterMax']
+
+        tokenMatrix.append(tokenIds)
+        positionMatrix1.append(positionValues1)
+        positionMatrix2.append(positionValues2)
+
+        labels.append(labelsMapping[label])
+
+
+
+    return np.array(labels, dtype='int32'), np.array(tokenMatrix, dtype='int32'), np.array(positionMatrix1, dtype='int32'), np.array(positionMatrix2, dtype='int32'),
+
+
+
+
+
+
+
+def load_data(embeddings, path_dataset):
+    files = [os.path.join(path_dataset, 'train.txt'), os.path.join(path_dataset, 'test.txt')]
+    for fileIdx in range(len(files)):
+        file = files[fileIdx]
+        for line in open(file):
+            splits = line.strip().split('\t')
+
+            label = splits[0]
+
+
+            sentence = splits[3]
+            tokens = sentence.split(" ")
+            maxSentenceLen[fileIdx] = max(maxSentenceLen[fileIdx], len(tokens))
+            for token in tokens:
+                words[token.lower()] = True
+
+
+    print("Max Sentence Lengths: ", maxSentenceLen)
+
+    # :: Read in word embeddings ::
+    # :: Read in word embeddings ::
+    word2Idx = embeddings.vocabulary.dic_words_ids
+    wordEmbeddings = embeddings.matrix
+
+
+    print("Embeddings shape: ", wordEmbeddings.shape)
+    print("Len words: ", len(words))
+
+
+
+    # :: Create token matrix ::
+    train_set = createTensor(files[0], word2Idx, max(maxSentenceLen))
+    test_set = createTensor(files[1], word2Idx, max(maxSentenceLen))
+
+
+    data = {'wordEmbeddings': wordEmbeddings, 'word2Idx': word2Idx,
+            'train_set': train_set, 'test_set': test_set}
+
+    return data
+
+
+
+    print("Data preprocessing done!")