Spaces:
Running
Running
Avijit Ghosh
commited on
Commit
·
981ea1d
1
Parent(s):
3d19330
removed notebook
Browse files- .gitignore +2 -0
- temp.ipynb +0 -568
.gitignore
CHANGED
@@ -1,2 +1,4 @@
|
|
1 |
temp/
|
2 |
.env
|
|
|
|
|
|
1 |
temp/
|
2 |
.env
|
3 |
+
*.ipynb
|
4 |
+
*.ipynb_checkpoints/
|
temp.ipynb
DELETED
@@ -1,568 +0,0 @@
|
|
1 |
-
{
|
2 |
-
"cells": [
|
3 |
-
{
|
4 |
-
"cell_type": "code",
|
5 |
-
"execution_count": 20,
|
6 |
-
"metadata": {},
|
7 |
-
"outputs": [],
|
8 |
-
"source": [
|
9 |
-
"import pandas as pd"
|
10 |
-
]
|
11 |
-
},
|
12 |
-
{
|
13 |
-
"cell_type": "code",
|
14 |
-
"execution_count": 21,
|
15 |
-
"metadata": {},
|
16 |
-
"outputs": [],
|
17 |
-
"source": [
|
18 |
-
"df = pd.read_csv('DemoData.csv')"
|
19 |
-
]
|
20 |
-
},
|
21 |
-
{
|
22 |
-
"cell_type": "code",
|
23 |
-
"execution_count": 22,
|
24 |
-
"metadata": {},
|
25 |
-
"outputs": [],
|
26 |
-
"source": [
|
27 |
-
"import pandas as pd\n",
|
28 |
-
"import yaml\n",
|
29 |
-
"import os\n",
|
30 |
-
"import ast\n",
|
31 |
-
"\n",
|
32 |
-
"# Create a folder to store YAML files if it doesn't exist\n",
|
33 |
-
"if not os.path.exists('configs'):\n",
|
34 |
-
" os.makedirs('configs')\n",
|
35 |
-
"\n",
|
36 |
-
"# Iterate over each row in the DataFrame\n",
|
37 |
-
"for index, row in df.iterrows():\n",
|
38 |
-
" # Extract Metaname and use it as the filename for YAML\n",
|
39 |
-
" filename = str(row['Metaname']) + '.yaml'\n",
|
40 |
-
" # Convert 'Screenshots' column to a Python list\n",
|
41 |
-
" screenshots_list = None\n",
|
42 |
-
" try:\n",
|
43 |
-
"\n",
|
44 |
-
" screenshots_list = ast.literal_eval(row['Screenshots'])\n",
|
45 |
-
" except:\n",
|
46 |
-
" screenshots_list = []\n",
|
47 |
-
" # Remove the 'Metaname' and 'Screenshots' columns from the data to be converted to YAML\n",
|
48 |
-
" row_data = row.drop(['Metaname', 'Screenshots'])\n",
|
49 |
-
" # Convert the remaining data to a dictionary\n",
|
50 |
-
" data_dict = row_data.to_dict()\n",
|
51 |
-
" # Add the 'Screenshots' list to the dictionary\n",
|
52 |
-
" data_dict['Screenshots'] = screenshots_list\n",
|
53 |
-
" # Write the data as YAML to a new file\n",
|
54 |
-
" with open(os.path.join('configs', filename), 'w') as yamlfile:\n",
|
55 |
-
" yaml.dump(data_dict, yamlfile, default_flow_style=False)"
|
56 |
-
]
|
57 |
-
},
|
58 |
-
{
|
59 |
-
"cell_type": "code",
|
60 |
-
"execution_count": 5,
|
61 |
-
"metadata": {},
|
62 |
-
"outputs": [
|
63 |
-
{
|
64 |
-
"data": {
|
65 |
-
"text/html": [
|
66 |
-
"<div>\n",
|
67 |
-
"<style scoped>\n",
|
68 |
-
" .dataframe tbody tr th:only-of-type {\n",
|
69 |
-
" vertical-align: middle;\n",
|
70 |
-
" }\n",
|
71 |
-
"\n",
|
72 |
-
" .dataframe tbody tr th {\n",
|
73 |
-
" vertical-align: top;\n",
|
74 |
-
" }\n",
|
75 |
-
"\n",
|
76 |
-
" .dataframe thead th {\n",
|
77 |
-
" text-align: right;\n",
|
78 |
-
" }\n",
|
79 |
-
"</style>\n",
|
80 |
-
"<table border=\"1\" class=\"dataframe\">\n",
|
81 |
-
" <thead>\n",
|
82 |
-
" <tr style=\"text-align: right;\">\n",
|
83 |
-
" <th></th>\n",
|
84 |
-
" <th>Group</th>\n",
|
85 |
-
" <th>Modality</th>\n",
|
86 |
-
" <th>Level</th>\n",
|
87 |
-
" <th>Metaname</th>\n",
|
88 |
-
" <th>Suggested Evaluation</th>\n",
|
89 |
-
" <th>What it is evaluating</th>\n",
|
90 |
-
" <th>Considerations</th>\n",
|
91 |
-
" <th>Link</th>\n",
|
92 |
-
" <th>URL</th>\n",
|
93 |
-
" <th>Screenshots</th>\n",
|
94 |
-
" <th>Applicable Models</th>\n",
|
95 |
-
" <th>Datasets</th>\n",
|
96 |
-
" <th>Hashtags</th>\n",
|
97 |
-
" </tr>\n",
|
98 |
-
" </thead>\n",
|
99 |
-
" <tbody>\n",
|
100 |
-
" <tr>\n",
|
101 |
-
" <th>0</th>\n",
|
102 |
-
" <td>BiasEvals</td>\n",
|
103 |
-
" <td>Text</td>\n",
|
104 |
-
" <td>Model</td>\n",
|
105 |
-
" <td>weat</td>\n",
|
106 |
-
" <td>Word Embedding Association Test (WEAT)</td>\n",
|
107 |
-
" <td>Associations and word embeddings based on Impl...</td>\n",
|
108 |
-
" <td>Although based in human associations, general ...</td>\n",
|
109 |
-
" <td>Semantics derived automatically from language ...</td>\n",
|
110 |
-
" <td>https://researchportal.bath.ac.uk/en/publicati...</td>\n",
|
111 |
-
" <td>['Images/WEAT1.png', 'Images/WEAT2.png']</td>\n",
|
112 |
-
" <td>NaN</td>\n",
|
113 |
-
" <td>NaN</td>\n",
|
114 |
-
" <td>NaN</td>\n",
|
115 |
-
" </tr>\n",
|
116 |
-
" <tr>\n",
|
117 |
-
" <th>1</th>\n",
|
118 |
-
" <td>BiasEvals</td>\n",
|
119 |
-
" <td>Text</td>\n",
|
120 |
-
" <td>Model</td>\n",
|
121 |
-
" <td>wefat</td>\n",
|
122 |
-
" <td>Word Embedding Factual As\\nsociation Test (WEFAT)</td>\n",
|
123 |
-
" <td>Associations and word embeddings based on Impl...</td>\n",
|
124 |
-
" <td>Although based in human associations, general ...</td>\n",
|
125 |
-
" <td>Semantics derived automatically from language ...</td>\n",
|
126 |
-
" <td>https://researchportal.bath.ac.uk/en/publicati...</td>\n",
|
127 |
-
" <td>NaN</td>\n",
|
128 |
-
" <td>NaN</td>\n",
|
129 |
-
" <td>NaN</td>\n",
|
130 |
-
" <td>NaN</td>\n",
|
131 |
-
" </tr>\n",
|
132 |
-
" <tr>\n",
|
133 |
-
" <th>2</th>\n",
|
134 |
-
" <td>BiasEvals</td>\n",
|
135 |
-
" <td>Text</td>\n",
|
136 |
-
" <td>Dataset</td>\n",
|
137 |
-
" <td>stereoset</td>\n",
|
138 |
-
" <td>StereoSet</td>\n",
|
139 |
-
" <td>Protected class stereotypes</td>\n",
|
140 |
-
" <td>Automating stereotype detection makes distingu...</td>\n",
|
141 |
-
" <td>StereoSet: Measuring stereotypical bias in pre...</td>\n",
|
142 |
-
" <td>https://arxiv.org/abs/2004.09456</td>\n",
|
143 |
-
" <td>NaN</td>\n",
|
144 |
-
" <td>NaN</td>\n",
|
145 |
-
" <td>NaN</td>\n",
|
146 |
-
" <td>NaN</td>\n",
|
147 |
-
" </tr>\n",
|
148 |
-
" <tr>\n",
|
149 |
-
" <th>3</th>\n",
|
150 |
-
" <td>BiasEvals</td>\n",
|
151 |
-
" <td>Text</td>\n",
|
152 |
-
" <td>Dataset</td>\n",
|
153 |
-
" <td>crwospairs</td>\n",
|
154 |
-
" <td>Crow-S Pairs</td>\n",
|
155 |
-
" <td>Protected class stereotypes</td>\n",
|
156 |
-
" <td>Automating stereotype detection makes distingu...</td>\n",
|
157 |
-
" <td>CrowS-Pairs: A Challenge Dataset for Measuring...</td>\n",
|
158 |
-
" <td>https://arxiv.org/abs/2010.00133</td>\n",
|
159 |
-
" <td>NaN</td>\n",
|
160 |
-
" <td>NaN</td>\n",
|
161 |
-
" <td>NaN</td>\n",
|
162 |
-
" <td>NaN</td>\n",
|
163 |
-
" </tr>\n",
|
164 |
-
" <tr>\n",
|
165 |
-
" <th>4</th>\n",
|
166 |
-
" <td>BiasEvals</td>\n",
|
167 |
-
" <td>Text</td>\n",
|
168 |
-
" <td>Output</td>\n",
|
169 |
-
" <td>honest</td>\n",
|
170 |
-
" <td>HONEST: Measuring Hurtful Sentence Completion ...</td>\n",
|
171 |
-
" <td>Protected class stereotypes and hurtful language</td>\n",
|
172 |
-
" <td>Automating stereotype detection makes distingu...</td>\n",
|
173 |
-
" <td>HONEST: Measuring Hurtful Sentence Completion ...</td>\n",
|
174 |
-
" <td>https://aclanthology.org/2021.naacl-main.191.pdf</td>\n",
|
175 |
-
" <td>NaN</td>\n",
|
176 |
-
" <td>NaN</td>\n",
|
177 |
-
" <td>NaN</td>\n",
|
178 |
-
" <td>NaN</td>\n",
|
179 |
-
" </tr>\n",
|
180 |
-
" <tr>\n",
|
181 |
-
" <th>5</th>\n",
|
182 |
-
" <td>BiasEvals</td>\n",
|
183 |
-
" <td>Image</td>\n",
|
184 |
-
" <td>Model</td>\n",
|
185 |
-
" <td>ieat</td>\n",
|
186 |
-
" <td>Image Embedding Association Test (iEAT)</td>\n",
|
187 |
-
" <td>Embedding associations</td>\n",
|
188 |
-
" <td>Although based in human associations, general ...</td>\n",
|
189 |
-
" <td>Image Representations Learned With Unsupervise...</td>\n",
|
190 |
-
" <td>https://dl.acm.org/doi/abs/10.1145/3442188.344...</td>\n",
|
191 |
-
" <td>NaN</td>\n",
|
192 |
-
" <td>NaN</td>\n",
|
193 |
-
" <td>NaN</td>\n",
|
194 |
-
" <td>NaN</td>\n",
|
195 |
-
" </tr>\n",
|
196 |
-
" <tr>\n",
|
197 |
-
" <th>6</th>\n",
|
198 |
-
" <td>BiasEvals</td>\n",
|
199 |
-
" <td>Image</td>\n",
|
200 |
-
" <td>Dataset</td>\n",
|
201 |
-
" <td>imagedataleak</td>\n",
|
202 |
-
" <td>Dataset leakage and model leakage</td>\n",
|
203 |
-
" <td>Gender and label bias</td>\n",
|
204 |
-
" <td>NaN</td>\n",
|
205 |
-
" <td>Balanced Datasets Are Not Enough: Estimating a...</td>\n",
|
206 |
-
" <td>https://arxiv.org/abs/1811.08489</td>\n",
|
207 |
-
" <td>NaN</td>\n",
|
208 |
-
" <td>NaN</td>\n",
|
209 |
-
" <td>NaN</td>\n",
|
210 |
-
" <td>NaN</td>\n",
|
211 |
-
" </tr>\n",
|
212 |
-
" <tr>\n",
|
213 |
-
" <th>7</th>\n",
|
214 |
-
" <td>BiasEvals</td>\n",
|
215 |
-
" <td>Image</td>\n",
|
216 |
-
" <td>Output</td>\n",
|
217 |
-
" <td>stablebias</td>\n",
|
218 |
-
" <td>Characterizing the variation in generated images</td>\n",
|
219 |
-
" <td>NaN</td>\n",
|
220 |
-
" <td>NaN</td>\n",
|
221 |
-
" <td>Stable bias: Analyzing societal representation...</td>\n",
|
222 |
-
" <td>https://arxiv.org/abs/2303.11408</td>\n",
|
223 |
-
" <td>NaN</td>\n",
|
224 |
-
" <td>NaN</td>\n",
|
225 |
-
" <td>NaN</td>\n",
|
226 |
-
" <td>NaN</td>\n",
|
227 |
-
" </tr>\n",
|
228 |
-
" <tr>\n",
|
229 |
-
" <th>8</th>\n",
|
230 |
-
" <td>BiasEvals</td>\n",
|
231 |
-
" <td>Image</td>\n",
|
232 |
-
" <td>Output</td>\n",
|
233 |
-
" <td>homoglyphbias</td>\n",
|
234 |
-
" <td>Effect of different scripts on text-to-image g...</td>\n",
|
235 |
-
" <td>It evaluates generated images for cultural ste...</td>\n",
|
236 |
-
" <td>NaN</td>\n",
|
237 |
-
" <td>Exploiting Cultural Biases via Homoglyphs in T...</td>\n",
|
238 |
-
" <td>https://arxiv.org/pdf/2209.08891.pdf</td>\n",
|
239 |
-
" <td>NaN</td>\n",
|
240 |
-
" <td>NaN</td>\n",
|
241 |
-
" <td>NaN</td>\n",
|
242 |
-
" <td>NaN</td>\n",
|
243 |
-
" </tr>\n",
|
244 |
-
" <tr>\n",
|
245 |
-
" <th>9</th>\n",
|
246 |
-
" <td>BiasEvals</td>\n",
|
247 |
-
" <td>Audio</td>\n",
|
248 |
-
" <td>Taxonomy (?)</td>\n",
|
249 |
-
" <td>notmyvoice</td>\n",
|
250 |
-
" <td>Not My Voice! A Taxonomy of Ethical and Safety...</td>\n",
|
251 |
-
" <td>Lists harms of audio/speech generators</td>\n",
|
252 |
-
" <td>Not necessarily evaluation but a good source o...</td>\n",
|
253 |
-
" <td>Not My Voice! A Taxonomy of Ethical and Safety...</td>\n",
|
254 |
-
" <td>https://arxiv.org/pdf/2402.01708.pdf</td>\n",
|
255 |
-
" <td>NaN</td>\n",
|
256 |
-
" <td>NaN</td>\n",
|
257 |
-
" <td>NaN</td>\n",
|
258 |
-
" <td>NaN</td>\n",
|
259 |
-
" </tr>\n",
|
260 |
-
" <tr>\n",
|
261 |
-
" <th>10</th>\n",
|
262 |
-
" <td>BiasEvals</td>\n",
|
263 |
-
" <td>Video</td>\n",
|
264 |
-
" <td>Output</td>\n",
|
265 |
-
" <td>videodiversemisinfo</td>\n",
|
266 |
-
" <td>Diverse Misinformation: Impacts of Human Biase...</td>\n",
|
267 |
-
" <td>Human led evaluations of deepfakes to understa...</td>\n",
|
268 |
-
" <td>Repr. harm, incite violence</td>\n",
|
269 |
-
" <td>Diverse Misinformation: Impacts of Human Biase...</td>\n",
|
270 |
-
" <td>https://arxiv.org/abs/2210.10026</td>\n",
|
271 |
-
" <td>NaN</td>\n",
|
272 |
-
" <td>NaN</td>\n",
|
273 |
-
" <td>NaN</td>\n",
|
274 |
-
" <td>NaN</td>\n",
|
275 |
-
" </tr>\n",
|
276 |
-
" <tr>\n",
|
277 |
-
" <th>11</th>\n",
|
278 |
-
" <td>Privacy</td>\n",
|
279 |
-
" <td>NaN</td>\n",
|
280 |
-
" <td>NaN</td>\n",
|
281 |
-
" <td>NaN</td>\n",
|
282 |
-
" <td>NaN</td>\n",
|
283 |
-
" <td>NaN</td>\n",
|
284 |
-
" <td>NaN</td>\n",
|
285 |
-
" <td>NaN</td>\n",
|
286 |
-
" <td>NaN</td>\n",
|
287 |
-
" <td>NaN</td>\n",
|
288 |
-
" <td>NaN</td>\n",
|
289 |
-
" <td>NaN</td>\n",
|
290 |
-
" <td>NaN</td>\n",
|
291 |
-
" </tr>\n",
|
292 |
-
" </tbody>\n",
|
293 |
-
"</table>\n",
|
294 |
-
"</div>"
|
295 |
-
],
|
296 |
-
"text/plain": [
|
297 |
-
" Group Modality Level Metaname \\\n",
|
298 |
-
"0 BiasEvals Text Model weat \n",
|
299 |
-
"1 BiasEvals Text Model wefat \n",
|
300 |
-
"2 BiasEvals Text Dataset stereoset \n",
|
301 |
-
"3 BiasEvals Text Dataset crwospairs \n",
|
302 |
-
"4 BiasEvals Text Output honest \n",
|
303 |
-
"5 BiasEvals Image Model ieat \n",
|
304 |
-
"6 BiasEvals Image Dataset imagedataleak \n",
|
305 |
-
"7 BiasEvals Image Output stablebias \n",
|
306 |
-
"8 BiasEvals Image Output homoglyphbias \n",
|
307 |
-
"9 BiasEvals Audio Taxonomy (?) notmyvoice \n",
|
308 |
-
"10 BiasEvals Video Output videodiversemisinfo \n",
|
309 |
-
"11 Privacy NaN NaN NaN \n",
|
310 |
-
"\n",
|
311 |
-
" Suggested Evaluation \\\n",
|
312 |
-
"0 Word Embedding Association Test (WEAT) \n",
|
313 |
-
"1 Word Embedding Factual As\\nsociation Test (WEFAT) \n",
|
314 |
-
"2 StereoSet \n",
|
315 |
-
"3 Crow-S Pairs \n",
|
316 |
-
"4 HONEST: Measuring Hurtful Sentence Completion ... \n",
|
317 |
-
"5 Image Embedding Association Test (iEAT) \n",
|
318 |
-
"6 Dataset leakage and model leakage \n",
|
319 |
-
"7 Characterizing the variation in generated images \n",
|
320 |
-
"8 Effect of different scripts on text-to-image g... \n",
|
321 |
-
"9 Not My Voice! A Taxonomy of Ethical and Safety... \n",
|
322 |
-
"10 Diverse Misinformation: Impacts of Human Biase... \n",
|
323 |
-
"11 NaN \n",
|
324 |
-
"\n",
|
325 |
-
" What it is evaluating \\\n",
|
326 |
-
"0 Associations and word embeddings based on Impl... \n",
|
327 |
-
"1 Associations and word embeddings based on Impl... \n",
|
328 |
-
"2 Protected class stereotypes \n",
|
329 |
-
"3 Protected class stereotypes \n",
|
330 |
-
"4 Protected class stereotypes and hurtful language \n",
|
331 |
-
"5 Embedding associations \n",
|
332 |
-
"6 Gender and label bias \n",
|
333 |
-
"7 NaN \n",
|
334 |
-
"8 It evaluates generated images for cultural ste... \n",
|
335 |
-
"9 Lists harms of audio/speech generators \n",
|
336 |
-
"10 Human led evaluations of deepfakes to understa... \n",
|
337 |
-
"11 NaN \n",
|
338 |
-
"\n",
|
339 |
-
" Considerations \\\n",
|
340 |
-
"0 Although based in human associations, general ... \n",
|
341 |
-
"1 Although based in human associations, general ... \n",
|
342 |
-
"2 Automating stereotype detection makes distingu... \n",
|
343 |
-
"3 Automating stereotype detection makes distingu... \n",
|
344 |
-
"4 Automating stereotype detection makes distingu... \n",
|
345 |
-
"5 Although based in human associations, general ... \n",
|
346 |
-
"6 NaN \n",
|
347 |
-
"7 NaN \n",
|
348 |
-
"8 NaN \n",
|
349 |
-
"9 Not necessarily evaluation but a good source o... \n",
|
350 |
-
"10 Repr. harm, incite violence \n",
|
351 |
-
"11 NaN \n",
|
352 |
-
"\n",
|
353 |
-
" Link \\\n",
|
354 |
-
"0 Semantics derived automatically from language ... \n",
|
355 |
-
"1 Semantics derived automatically from language ... \n",
|
356 |
-
"2 StereoSet: Measuring stereotypical bias in pre... \n",
|
357 |
-
"3 CrowS-Pairs: A Challenge Dataset for Measuring... \n",
|
358 |
-
"4 HONEST: Measuring Hurtful Sentence Completion ... \n",
|
359 |
-
"5 Image Representations Learned With Unsupervise... \n",
|
360 |
-
"6 Balanced Datasets Are Not Enough: Estimating a... \n",
|
361 |
-
"7 Stable bias: Analyzing societal representation... \n",
|
362 |
-
"8 Exploiting Cultural Biases via Homoglyphs in T... \n",
|
363 |
-
"9 Not My Voice! A Taxonomy of Ethical and Safety... \n",
|
364 |
-
"10 Diverse Misinformation: Impacts of Human Biase... \n",
|
365 |
-
"11 NaN \n",
|
366 |
-
"\n",
|
367 |
-
" URL \\\n",
|
368 |
-
"0 https://researchportal.bath.ac.uk/en/publicati... \n",
|
369 |
-
"1 https://researchportal.bath.ac.uk/en/publicati... \n",
|
370 |
-
"2 https://arxiv.org/abs/2004.09456 \n",
|
371 |
-
"3 https://arxiv.org/abs/2010.00133 \n",
|
372 |
-
"4 https://aclanthology.org/2021.naacl-main.191.pdf \n",
|
373 |
-
"5 https://dl.acm.org/doi/abs/10.1145/3442188.344... \n",
|
374 |
-
"6 https://arxiv.org/abs/1811.08489 \n",
|
375 |
-
"7 https://arxiv.org/abs/2303.11408 \n",
|
376 |
-
"8 https://arxiv.org/pdf/2209.08891.pdf \n",
|
377 |
-
"9 https://arxiv.org/pdf/2402.01708.pdf \n",
|
378 |
-
"10 https://arxiv.org/abs/2210.10026 \n",
|
379 |
-
"11 NaN \n",
|
380 |
-
"\n",
|
381 |
-
" Screenshots Applicable Models Datasets \\\n",
|
382 |
-
"0 ['Images/WEAT1.png', 'Images/WEAT2.png'] NaN NaN \n",
|
383 |
-
"1 NaN NaN NaN \n",
|
384 |
-
"2 NaN NaN NaN \n",
|
385 |
-
"3 NaN NaN NaN \n",
|
386 |
-
"4 NaN NaN NaN \n",
|
387 |
-
"5 NaN NaN NaN \n",
|
388 |
-
"6 NaN NaN NaN \n",
|
389 |
-
"7 NaN NaN NaN \n",
|
390 |
-
"8 NaN NaN NaN \n",
|
391 |
-
"9 NaN NaN NaN \n",
|
392 |
-
"10 NaN NaN NaN \n",
|
393 |
-
"11 NaN NaN NaN \n",
|
394 |
-
"\n",
|
395 |
-
" Hashtags \n",
|
396 |
-
"0 NaN \n",
|
397 |
-
"1 NaN \n",
|
398 |
-
"2 NaN \n",
|
399 |
-
"3 NaN \n",
|
400 |
-
"4 NaN \n",
|
401 |
-
"5 NaN \n",
|
402 |
-
"6 NaN \n",
|
403 |
-
"7 NaN \n",
|
404 |
-
"8 NaN \n",
|
405 |
-
"9 NaN \n",
|
406 |
-
"10 NaN \n",
|
407 |
-
"11 NaN "
|
408 |
-
]
|
409 |
-
},
|
410 |
-
"execution_count": 5,
|
411 |
-
"metadata": {},
|
412 |
-
"output_type": "execute_result"
|
413 |
-
}
|
414 |
-
],
|
415 |
-
"source": [
|
416 |
-
"df"
|
417 |
-
]
|
418 |
-
},
|
419 |
-
{
|
420 |
-
"cell_type": "code",
|
421 |
-
"execution_count": 9,
|
422 |
-
"metadata": {},
|
423 |
-
"outputs": [],
|
424 |
-
"source": [
|
425 |
-
"import urllib.request\n",
|
426 |
-
"from bs4 import BeautifulSoup\n",
|
427 |
-
"\n",
|
428 |
-
"from pypdf import PdfReader \n",
|
429 |
-
"from urllib.request import urlretrieve\n",
|
430 |
-
"\n",
|
431 |
-
"import pdfplumber\n",
|
432 |
-
"\n"
|
433 |
-
]
|
434 |
-
},
|
435 |
-
{
|
436 |
-
"cell_type": "code",
|
437 |
-
"execution_count": 12,
|
438 |
-
"metadata": {},
|
439 |
-
"outputs": [
|
440 |
-
{
|
441 |
-
"name": "stdout",
|
442 |
-
"output_type": "stream",
|
443 |
-
"text": [
|
444 |
-
"https://researchportal.bath.ac.uk/en/publications/semantics-derived-automatically-from-language-corpora-necessarily\n",
|
445 |
-
"\n",
|
446 |
-
" Semantics derived automatically from language corpora contain human-like biases\n",
|
447 |
-
" — the University of Bath's research portal\n",
|
448 |
-
"https://researchportal.bath.ac.uk/en/publications/semantics-derived-automatically-from-language-corpora-necessarily\n",
|
449 |
-
"\n",
|
450 |
-
" Semantics derived automatically from language corpora contain human-like biases\n",
|
451 |
-
" — the University of Bath's research portal\n",
|
452 |
-
"https://arxiv.org/abs/1903.10561\n",
|
453 |
-
"[1903.10561] On Measuring Social Biases in Sentence Encoders\n",
|
454 |
-
"https://dl.acm.org/doi/abs/10.5555/3454287.3455472\n",
|
455 |
-
"Error\n",
|
456 |
-
"https://arxiv.org/abs/2004.09456\n",
|
457 |
-
"[2004.09456] StereoSet: Measuring stereotypical bias in pretrained language models\n",
|
458 |
-
"https://arxiv.org/abs/2010.00133\n",
|
459 |
-
"[2010.00133] CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models\n",
|
460 |
-
"https://aclanthology.org/2021.naacl-main.191.pdf\n"
|
461 |
-
]
|
462 |
-
},
|
463 |
-
{
|
464 |
-
"name": "stderr",
|
465 |
-
"output_type": "stream",
|
466 |
-
"text": [
|
467 |
-
"Some characters could not be decoded, and were replaced with REPLACEMENT CHARACTER.\n"
|
468 |
-
]
|
469 |
-
},
|
470 |
-
{
|
471 |
-
"name": "stdout",
|
472 |
-
"output_type": "stream",
|
473 |
-
"text": [
|
474 |
-
"HONEST: Measuring Hurtful Sentence Completion in Language Models\n",
|
475 |
-
"nan\n",
|
476 |
-
"Error\n",
|
477 |
-
"https://aclanthology.org/2022.findings-acl.165.pdf\n"
|
478 |
-
]
|
479 |
-
},
|
480 |
-
{
|
481 |
-
"name": "stderr",
|
482 |
-
"output_type": "stream",
|
483 |
-
"text": [
|
484 |
-
"Some characters could not be decoded, and were replaced with REPLACEMENT CHARACTER.\n"
|
485 |
-
]
|
486 |
-
},
|
487 |
-
{
|
488 |
-
"name": "stdout",
|
489 |
-
"output_type": "stream",
|
490 |
-
"text": [
|
491 |
-
"BBQ: A Hand-Built Bias Benchmark for Question Answering \n",
|
492 |
-
"https://aclanthology.org/2022.findings-naacl.42.pdf\n"
|
493 |
-
]
|
494 |
-
},
|
495 |
-
{
|
496 |
-
"name": "stderr",
|
497 |
-
"output_type": "stream",
|
498 |
-
"text": [
|
499 |
-
"Some characters could not be decoded, and were replaced with REPLACEMENT CHARACTER.\n"
|
500 |
-
]
|
501 |
-
},
|
502 |
-
{
|
503 |
-
"name": "stdout",
|
504 |
-
"output_type": "stream",
|
505 |
-
"text": [
|
506 |
-
"On Measuring Social Biases in Prompt-Based Multi-Task Learning\n"
|
507 |
-
]
|
508 |
-
}
|
509 |
-
],
|
510 |
-
"source": [
|
511 |
-
"def get_page_title(url):\n",
|
512 |
-
" soup = BeautifulSoup(urllib.request.urlopen(url))\n",
|
513 |
-
" return soup.title.string\n",
|
514 |
-
"\n",
|
515 |
-
"\n",
|
516 |
-
"def extract_pdf_title(url):\n",
|
517 |
-
" urlretrieve(url, 'temp.pdf')\n",
|
518 |
-
" with pdfplumber.open('temp.pdf') as pdf:\n",
|
519 |
-
" for page in pdf.pages:\n",
|
520 |
-
" for line in page.extract_text().split('\\n'):\n",
|
521 |
-
" return line\n",
|
522 |
-
" return \"\"\n",
|
523 |
-
"\n",
|
524 |
-
" \n",
|
525 |
-
" \n",
|
526 |
-
"for url in df['URL'][:10]:\n",
|
527 |
-
" try:\n",
|
528 |
-
" print(url)\n",
|
529 |
-
" title = get_page_title(url)\n",
|
530 |
-
" print(title)\n",
|
531 |
-
" except:\n",
|
532 |
-
" try:\n",
|
533 |
-
" title = extract_pdf_title(url)\n",
|
534 |
-
" print(title)\n",
|
535 |
-
" except:\n",
|
536 |
-
" print(\"Error\")"
|
537 |
-
]
|
538 |
-
},
|
539 |
-
{
|
540 |
-
"cell_type": "code",
|
541 |
-
"execution_count": null,
|
542 |
-
"metadata": {},
|
543 |
-
"outputs": [],
|
544 |
-
"source": []
|
545 |
-
}
|
546 |
-
],
|
547 |
-
"metadata": {
|
548 |
-
"kernelspec": {
|
549 |
-
"display_name": "gradio",
|
550 |
-
"language": "python",
|
551 |
-
"name": "python3"
|
552 |
-
},
|
553 |
-
"language_info": {
|
554 |
-
"codemirror_mode": {
|
555 |
-
"name": "ipython",
|
556 |
-
"version": 3
|
557 |
-
},
|
558 |
-
"file_extension": ".py",
|
559 |
-
"mimetype": "text/x-python",
|
560 |
-
"name": "python",
|
561 |
-
"nbconvert_exporter": "python",
|
562 |
-
"pygments_lexer": "ipython3",
|
563 |
-
"version": "3.12.2"
|
564 |
-
}
|
565 |
-
},
|
566 |
-
"nbformat": 4,
|
567 |
-
"nbformat_minor": 2
|
568 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|