v2 filter out link texts that are dictionary words

.gitignore

@@ -0,0 +1 @@
+.venv/

2018thresh10dev.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53c34495315acde41ca549eaa1ad02726e79c6f2aad46b77f84b11fb2459e666
-size 55066171
+oid sha256:6690985884c8b4a58d0377ef2152bb1fbb233355df68d662539da856e697a414
+size 51406267

2018thresh10test.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8089f88487de91ed778ecfed210f0f13488e3f670404c210ebd72ee3f1b268de
-size 36685709
+oid sha256:87ba47d64b3223a9ceda6c4c414312c4c8fa412c023d63ba8495909206f908fd
+size 34147126

2018thresh10train.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2618a03dd0278017b987759f723cc7525d19ddc3c79247a187bcf0a2601663ed
-size 274277317
+oid sha256:6a4c19706eb30b58641b06fa0e5efe53cf545f6bf49e3b5a30e62d96243517c8
+size 255547715

2018thresh20dev.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:06c9b05cc92746020c8379d55d8738ca6923129ca8a15766df1687625d73cb30
-size 39942761
+oid sha256:e80d37c92672e2b2041d47035c737db5c7dc0a9c5938ede28c46cdc25321be5a
+size 37218821

2018thresh20test.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:feea8c3d46deb539e8004bc5193a15509cb3b352d93431dbe5131a5e4356b9d9
-size 26669268
+oid sha256:7204244b6dc46350616835b741f925b98f50f6e9b4d5585d7815f226d5afde95
+size 24867451

2018thresh20train.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ebd8c98658d16de99412d9f52d3b98409c0bcefd36409378fe303d5dd61c9eba
-size 198305773
+oid sha256:9fcda4ee523a653d09f34c72e2134d56466d145c56a95596468ca7a1f4f9bf9c
+size 185091962

2018thresh5dev.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf2e6b8bf0b091f3cbcc3cfbdd831989e35be1f8331de0a2f0d217e574c189c4
-size 71983269
+oid sha256:abf0f45844c3aa91747be22502d9d91570365aa0d085a1faaec17a04a84028c5
+size 67296490

2018thresh5test.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f654c09ae5f8ffbb80ddcb79c2be63e37b8ae184ef99c9555fb1c1625fedf1b7
-size 48195090
+oid sha256:7d9c39063394851308a52c71de4cae38386bfe7e16f73a355221dafe46a74b0f
+size 44759191

2018thresh5train.csv

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:866bad7a6fc1fcd477f87be9b856d289e47547878fcaeb000d1dba768a7b7468
-size 359260993
+oid sha256:e7b71f64c53d1aea33ec8bd6067b99c766625897904fd7cccdd82d60eac17b5c
+size 334315073

generate_wes_data.py

@@ -1,14 +1,19 @@
 from datasets import load_dataset
 import pandas as pd
+from nltk.corpus import words
+from nltk import WordNetLemmatizer
 import numpy as np
 from tqdm import tqdm
 
 from collections import defaultdict
 from operator import itemgetter as ig
 from itertools import islice, chain, repeat
-from random import sample, choice, shuffle
+from random import seed, sample, choice, shuffle
 from gc import collect
 
+filter_dict = set(words.words())
+ltize = WordNetLemmatizer().lemmatize
+
 def generate_splits(subset, split=[0.75, 0.15, 0.1]):
     assert abs(sum(split) - 1.0) < 0.0001
     # get the data in dictionary form
@@ -16,6 +21,9 @@ def generate_splits(subset, split=[0.75, 0.15, 0.1]):
     ds = load_dataset('Exr0n/wiki-entity-similarity', subset, split='train')
     ds = list(tqdm(ds, total=len(ds)))
     for article, link in tqdm(map(ig('article', 'link_text'), ds), total=len(ds)):
+        if (ltize(article.lower()) not in filter_dict) and (ltize(link.lower()) in filter_dict):
+            # print(article, link, 'not quite right!')
+            continue  # remove if link text is a dictionary word but article is not
         groups[article].append(link)
     del ds
 
@@ -38,10 +46,10 @@ def generate_splits(subset, split=[0.75, 0.15, 0.1]):
     for i, keys in enumerate(splits):
         for key in keys:
             try:
-                got = sample(keys, len(groups[key])+1)
+                got = sample(keys, len(groups[key])+1)  # sample n+1 keys
                 ret[i].append(
-                    [(key, choice(groups[k])) for k in got if k != key]
-                    [:len(groups[key])]
+                    [(key, choice(groups[k])) for k in got if k != key] # get a random link title from that key, if it's not the current key
+                    [:len(groups[key])] # ensure we don't have too many
                 )
             except ValueError:
                 raise ValueError("well frick one group is bigger than all the others combined. try sampling one at a time")
@@ -51,14 +59,16 @@ def generate_splits(subset, split=[0.75, 0.15, 0.1]):
 
 
 if __name__ == '__main__':
+    seed(0x326ccc)
+    year = 2018
     for size in [5, 10, 20]:
-        x = generate_splits(subset='2018thresh' + str(size) + 'corpus')
+        x = generate_splits(subset=f'{year}thresh' + str(size) + 'corpus')
 
         for (data, labels), split in zip(x, ['train', 'dev', 'test']):
             articles, lts = list(zip(*data))
             df = pd.DataFrame({ 'article': articles, 'link_text': lts, 'is_same': list(labels) })
             df = df.sample(frac=1).reset_index(drop=True)
-            df.to_csv('2018thresh' + str(size) + split + '.csv', index=False)
+            df.to_csv(f'{year}thresh' + str(size) + split + '.csv', index=False)
             # print(df.head(30), df.tail(30))
 
     # tests