add arxiv link
Browse files- data/tweet_qa/test.jsonl +2 -2
- data/tweet_qa/train.jsonl +2 -2
- data/tweet_qa/validation.jsonl +2 -2
- process/tweet_qa.py +2 -0
- super_tweeteval.py +10 -4
data/tweet_qa/test.jsonl
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:febf2fc5ae0ab149711709bf6a00ee5a96ccd0b5d00dac4dc50234a2621c0989
|
3 |
+
size 322894
|
data/tweet_qa/train.jsonl
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:662ffee064429bd16dd0ca1a72d8e810066eb30ad864b99c700e3d289722748f
|
3 |
+
size 2519331
|
data/tweet_qa/validation.jsonl
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:41478f2ae04546e6f6aa869fb947395a444e596c1923e9ddfc3312e6886bb412
|
3 |
+
size 285505
|
process/tweet_qa.py
CHANGED
@@ -11,6 +11,8 @@ def process(tmp):
|
|
11 |
for i in tmp:
|
12 |
i['text'] = i.pop('paragraph_question')
|
13 |
i['gold_label_str'] = i.pop('answer')
|
|
|
|
|
14 |
return tmp
|
15 |
|
16 |
train = process(data["train"].to_pandas())
|
|
|
11 |
for i in tmp:
|
12 |
i['text'] = i.pop('paragraph_question')
|
13 |
i['gold_label_str'] = i.pop('answer')
|
14 |
+
i.pop("paragraph")
|
15 |
+
i.pop("question")
|
16 |
return tmp
|
17 |
|
18 |
train = process(data["train"].to_pandas())
|
super_tweeteval.py
CHANGED
@@ -2,7 +2,7 @@
|
|
2 |
import json
|
3 |
import datasets
|
4 |
|
5 |
-
_VERSION = "0.1.
|
6 |
_SUPER_TWEETEVAL_CITATION = """TBA"""
|
7 |
_SUPER_TWEETEVAL_DESCRIPTION = """TBA"""
|
8 |
_TWEET_TOPIC_DESCRIPTION = """
|
@@ -199,9 +199,16 @@ class SuperTweetEval(datasets.GeneratorBasedBuilder):
|
|
199 |
name="tweet_qa",
|
200 |
description=_TWEET_QA_DESCRIPTION,
|
201 |
citation=_TWEET_QA_CITATION,
|
202 |
-
features=["text", "gold_label_str"
|
203 |
data_url="https://huggingface.co/datasets/cardiffnlp/super_tweet_eval/resolve/main/data/tweet_qa",
|
204 |
),
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
205 |
SuperTweetEvalConfig(
|
206 |
name="tweet_intimacy",
|
207 |
description=_TWEET_INTIMACY_DESCRIPTION,
|
@@ -284,7 +291,7 @@ class SuperTweetEval(datasets.GeneratorBasedBuilder):
|
|
284 |
features['date_2'] = datasets.Value("string")
|
285 |
if self.config.name == "tweet_hate":
|
286 |
label_classes = [
|
287 |
-
'hate_gender','hate_race', 'hate_sexuality', 'hate_religion','hate_origin', 'hate_disability',
|
288 |
'hate_age', 'not_hate']
|
289 |
features['gold_label'] = datasets.features.ClassLabel(names=label_classes)
|
290 |
features["text"] = datasets.Value("string")
|
@@ -310,7 +317,6 @@ class SuperTweetEval(datasets.GeneratorBasedBuilder):
|
|
310 |
features["text"] = datasets.Value("string")
|
311 |
features["target"] = datasets.Value("string")
|
312 |
|
313 |
-
|
314 |
return datasets.DatasetInfo(
|
315 |
description=_SUPER_TWEETEVAL_DESCRIPTION + "\n" + self.config.description,
|
316 |
features=datasets.Features(features),
|
|
|
2 |
import json
|
3 |
import datasets
|
4 |
|
5 |
+
_VERSION = "0.1.42"
|
6 |
_SUPER_TWEETEVAL_CITATION = """TBA"""
|
7 |
_SUPER_TWEETEVAL_DESCRIPTION = """TBA"""
|
8 |
_TWEET_TOPIC_DESCRIPTION = """
|
|
|
199 |
name="tweet_qa",
|
200 |
description=_TWEET_QA_DESCRIPTION,
|
201 |
citation=_TWEET_QA_CITATION,
|
202 |
+
features=["text", "gold_label_str"],
|
203 |
data_url="https://huggingface.co/datasets/cardiffnlp/super_tweet_eval/resolve/main/data/tweet_qa",
|
204 |
),
|
205 |
+
SuperTweetEvalConfig(
|
206 |
+
name="tweet_qg",
|
207 |
+
description=_TWEET_QA_DESCRIPTION,
|
208 |
+
citation=_TWEET_QA_CITATION,
|
209 |
+
features=["text", "gold_label_str"],
|
210 |
+
data_url="https://huggingface.co/datasets/cardiffnlp/super_tweet_eval/resolve/main/data/tweet_qg",
|
211 |
+
),
|
212 |
SuperTweetEvalConfig(
|
213 |
name="tweet_intimacy",
|
214 |
description=_TWEET_INTIMACY_DESCRIPTION,
|
|
|
291 |
features['date_2'] = datasets.Value("string")
|
292 |
if self.config.name == "tweet_hate":
|
293 |
label_classes = [
|
294 |
+
'hate_gender', 'hate_race', 'hate_sexuality', 'hate_religion','hate_origin', 'hate_disability',
|
295 |
'hate_age', 'not_hate']
|
296 |
features['gold_label'] = datasets.features.ClassLabel(names=label_classes)
|
297 |
features["text"] = datasets.Value("string")
|
|
|
317 |
features["text"] = datasets.Value("string")
|
318 |
features["target"] = datasets.Value("string")
|
319 |
|
|
|
320 |
return datasets.DatasetInfo(
|
321 |
description=_SUPER_TWEETEVAL_DESCRIPTION + "\n" + self.config.description,
|
322 |
features=datasets.Features(features),
|