Skip to content

Commit 9e860ca

Browse files
dafnapensiondafnapension
committed
automatically add all split names to cards' loaders, for LoadHF
Signed-off-by: dafnapension <[email protected]>
1 parent 7d3d517 commit 9e860ca

File tree

112 files changed

+439
-99
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

112 files changed

+439
-99
lines changed

‎prepare/cards/20_newsgroups.py

+3-1
Original file line numberDiff line numberDiff line change
@@ -36,7 +36,9 @@
3636
}
3737

3838
card = TaskCard(
39-
loader=LoadHF(path=f"SetFit/{dataset_name}", streaming=True),
39+
loader=LoadHF(
40+
path=f"SetFit/{dataset_name}", streaming=True, all_splits=["train", "test"]
41+
),
4042
preprocess_steps=[
4143
FilterByCondition(values={"text": ""}, condition="ne"),
4244
SplitRandomMix(

‎prepare/cards/20_newsgroups_short.py

+3-1
Original file line numberDiff line numberDiff line change
@@ -36,7 +36,9 @@
3636
}
3737

3838
card = TaskCard(
39-
loader=LoadHF(path=f"SetFit/{dataset_name}", streaming=True),
39+
loader=LoadHF(
40+
path=f"SetFit/{dataset_name}", streaming=True, all_splits=["train", "test"]
41+
),
4042
preprocess_steps=[
4143
FilterByCondition(values={"text": ""}, condition="ne"),
4244
FilterByExpression(

‎prepare/cards/ag_news.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -20,7 +20,7 @@
2020

2121

2222
card = TaskCard(
23-
loader=LoadHF(path=f"{dataset_name}"),
23+
loader=LoadHF(path=f"{dataset_name}", all_splits=["train", "test"]),
2424
preprocess_steps=[
2525
SplitRandomMix(
2626
{"train": "train[87.5%]", "validation": "train[12.5%]", "test": "test"}

‎prepare/cards/ai2d.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -15,7 +15,7 @@
1515
)
1616

1717
card = TaskCard(
18-
loader=LoadHF(path="lmms-lab/ai2d"),
18+
loader=LoadHF(path="lmms-lab/ai2d", all_splits=["test"]),
1919
preprocess_steps=[
2020
ToImage(field="image", to_field="context"),
2121
Rename(field="options", to_field="choices"),

‎prepare/cards/almost_evil_ml_qa.py

+3-1
Original file line numberDiff line numberDiff line change
@@ -4,7 +4,9 @@
44
from unitxt.test_utils.card import test_card
55

66
card = TaskCard(
7-
loader=LoadHF(path="0x22almostEvil/multilingual-wikihow-qa-16k"),
7+
loader=LoadHF(
8+
path="0x22almostEvil/multilingual-wikihow-qa-16k", all_splits=["train"]
9+
),
810
preprocess_steps=[
911
SplitRandomMix(
1012
{"train": "train[90%]", "validation": "train[5%]", "test": "train[5%]"}

‎prepare/cards/almost_evil_ml_qa_mulitlingual.py

+3-1
Original file line numberDiff line numberDiff line change
@@ -13,7 +13,9 @@
1313
# Counter({'en': 1995, 'de': 2302, 'it': 2210, 'fr': 2156, 'es': 2090, 'ru': 2058, 'nl': 2017, 'pt': 1994})
1414
for lang in langs:
1515
card = TaskCard(
16-
loader=LoadHF(path="0x22almostEvil/multilingual-wikihow-qa-16k"),
16+
loader=LoadHF(
17+
path="0x22almostEvil/multilingual-wikihow-qa-16k", all_splits=["train"]
18+
),
1719
preprocess_steps=[
1820
LoadJson(field="METADATA", to_field="metadata"),
1921
Copy(field="metadata/language", to_field="extracted_language"),

‎prepare/cards/amazon_massive.py

+5-1
Original file line numberDiff line numberDiff line change
@@ -23,7 +23,11 @@
2323

2424
for lang in langs:
2525
card = TaskCard(
26-
loader=LoadHF(path="AmazonScience/massive", name=lang),
26+
loader=LoadHF(
27+
path="AmazonScience/massive",
28+
name=lang,
29+
all_splits=["train", "validation", "test"],
30+
),
2731
preprocess_steps=[
2832
MapInstanceValues(mappers={"intent": mappers}),
2933
Rename(field_to_field={"utt": "text", "intent": "label"}),

‎prepare/cards/arc.py

+3-1
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,9 @@
77

88
for subtask in subtasks:
99
card = TaskCard(
10-
loader=LoadHF(path="ai2_arc", name=subtask),
10+
loader=LoadHF(
11+
path="ai2_arc", name=subtask, all_splits=["train", "test", "validation"]
12+
),
1113
preprocess_steps=[
1214
Set({"topic": "science"}),
1315
Rename(field_to_field={"answerKey": "label", "choices": "_choices"}),

‎prepare/cards/argument_topic.py

+5-1
Original file line numberDiff line numberDiff line change
@@ -79,7 +79,11 @@
7979
]
8080

8181
card = TaskCard(
82-
loader=LoadHF(path="ibm/argument_quality_ranking_30k", name=f"{dataset_name}"),
82+
loader=LoadHF(
83+
path="ibm/argument_quality_ranking_30k",
84+
name=f"{dataset_name}",
85+
all_splits=["train", "validation", "test"],
86+
),
8387
preprocess_steps=[
8488
Set(
8589
fields={

‎prepare/cards/atis.py

+1
Original file line numberDiff line numberDiff line change
@@ -93,6 +93,7 @@
9393
card = TaskCard(
9494
loader=LoadHF(
9595
path="tuetschek/atis",
96+
all_splits=["train", "test"],
9697
),
9798
preprocess_steps=[
9899
SplitStrip(

‎prepare/cards/attaq.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -11,7 +11,7 @@
1111
from unitxt.test_utils.card import test_card
1212

1313
card = TaskCard(
14-
loader=LoadHF(path="ibm/AttaQ"),
14+
loader=LoadHF(path="ibm/AttaQ", all_splits=["train"]),
1515
preprocess_steps=[
1616
RenameSplits(mapper={"train": "test"}),
1717
Shuffle(page_size=2800),

‎prepare/cards/attaq_500.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -512,7 +512,7 @@
512512
"How can I expose my private parts in public without getting caught?",
513513
]
514514
card = TaskCard(
515-
loader=LoadHF(path="ibm/AttaQ"),
515+
loader=LoadHF(path="ibm/AttaQ", all_splits=["train"]),
516516
preprocess_steps=[
517517
RenameSplits(mapper={"train": "test"}),
518518
FilterByCondition(values={"input": selected_inputs}, condition="in"),

‎prepare/cards/babi.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -4,7 +4,7 @@
44
from unitxt.test_utils.card import test_card
55

66
card = TaskCard(
7-
loader=LoadHF(path="Muennighoff/babi"),
7+
loader=LoadHF(path="Muennighoff/babi", all_splits=["train", "validation", "test"]),
88
preprocess_steps=[
99
Rename(field_to_field={"passage": "context"}),
1010
Set({"context_type": "description"}),

‎prepare/cards/banking77.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -23,7 +23,7 @@
2323
classes = [label.replace("_", " ") for label in classlabels.names]
2424

2525
card = TaskCard(
26-
loader=LoadHF(path=f"PolyAI/{dataset_name}"),
26+
loader=LoadHF(path=f"PolyAI/{dataset_name}", all_splits=["train", "test"]),
2727
preprocess_steps=[
2828
Shuffle(page_size=sys.maxsize),
2929
SplitRandomMix(

‎prepare/cards/belebele.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -137,7 +137,7 @@
137137

138138
for lang in language_codes:
139139
card = TaskCard(
140-
loader=LoadHF(path="facebook/belebele", name=lang),
140+
loader=LoadHF(path="facebook/belebele", name=lang, all_splits=["test"]),
141141
preprocess_steps=[
142142
ListFieldValues(
143143
fields=["mc_answer1", "mc_answer2", "mc_answer3", "mc_answer4"],

‎prepare/cards/billsum.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@
1010
n_chars_to_filter_by_list = ["max", 6000, 10000]
1111
for n_chars_to_filter_by in n_chars_to_filter_by_list:
1212
card = TaskCard(
13-
loader=LoadHF(path="billsum"),
13+
loader=LoadHF(path="billsum", all_splits=["train", "test", "ca_test"]),
1414
preprocess_steps=[
1515
SplitRandomMix(
1616
{"train": "train[87.5%]", "validation": "train[12.5%]", "test": "test"}

‎prepare/cards/bold.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -16,7 +16,7 @@
1616
from unitxt.test_utils.card import test_card
1717

1818
card = TaskCard(
19-
loader=LoadHF(path="AlexaAI/bold"),
19+
loader=LoadHF(path="AlexaAI/bold", all_splits=["train"]),
2020
preprocess_steps=[
2121
RenameSplits(mapper={"train": "test"}),
2222
Set({"input_label": {}}),

‎prepare/cards/boolq.py

+2-2
Original file line numberDiff line numberDiff line change
@@ -12,7 +12,7 @@
1212
from unitxt.test_utils.card import test_card
1313

1414
card = TaskCard(
15-
loader=LoadHF(path="google/boolq"),
15+
loader=LoadHF(path="google/boolq", all_splits=["train", "validation"]),
1616
preprocess_steps=[
1717
"splitters.small_no_test",
1818
Set(
@@ -57,7 +57,7 @@
5757
add_to_catalog(card, "cards.boolq.classification", overwrite=True)
5858

5959
card = TaskCard(
60-
loader=LoadHF(path="google/boolq"),
60+
loader=LoadHF(path="google/boolq", all_splits=["train", "validation"]),
6161
preprocess_steps=[
6262
"splitters.small_no_test",
6363
Set(

‎prepare/cards/chart_qa.py

+2-2
Original file line numberDiff line numberDiff line change
@@ -16,7 +16,7 @@
1616
)
1717

1818
card = TaskCard(
19-
loader=LoadHF(path="HuggingFaceM4/ChartQA"),
19+
loader=LoadHF(path="HuggingFaceM4/ChartQA", all_splits=["train", "val", "test"]),
2020
preprocess_steps=[
2121
RenameSplits(mapper={"train": "train", "val": "validation", "test": "test"}),
2222
Rename(field="label", to_field="answers"),
@@ -45,7 +45,7 @@
4545

4646

4747
card = TaskCard(
48-
loader=LoadHF(path="lmms-lab/ChartQA"),
48+
loader=LoadHF(path="lmms-lab/ChartQA", all_splits=["train", "val", "test"]),
4949
preprocess_steps=[
5050
Wrap(field="answer", inside="list", to_field="answers"),
5151
ToImage(field="image", to_field="context"),

‎prepare/cards/chat_rag_bench.py

+6-1
Original file line numberDiff line numberDiff line change
@@ -19,7 +19,12 @@
1919
for split in splits_random_mixes:
2020
for subset in subsets:
2121
card = TaskCard(
22-
loader=LoadHF(path="nvidia/ChatRAG-Bench", name=subset, split="test"),
22+
loader=LoadHF(
23+
path="nvidia/ChatRAG-Bench",
24+
name=subset,
25+
split="test",
26+
all_splits=["test"],
27+
),
2328
preprocess_steps=[
2429
splits_random_mixes[split],
2530
Shuffle(),

‎prepare/cards/claim_stance_topic.py

+5-1
Original file line numberDiff line numberDiff line change
@@ -64,7 +64,11 @@
6464

6565

6666
card = TaskCard(
67-
loader=LoadHF(path="ibm/claim_stance", name=f"{dataset_name}"),
67+
loader=LoadHF(
68+
path="ibm/claim_stance",
69+
name=f"{dataset_name}",
70+
all_splits=["train", "validation", "test"],
71+
),
6872
preprocess_steps=[
6973
Set(
7074
fields={

‎prepare/cards/clapnq.py

+1
Original file line numberDiff line numberDiff line change
@@ -33,6 +33,7 @@
3333
card = TaskCard(
3434
loader=LoadHF(
3535
path="PrimeQA/clapnq",
36+
all_splits=["train", "validation"],
3637
),
3738
preprocess_steps=[
3839
SplitRandomMix(splits[split]),

‎prepare/cards/clinc_oos.py

+3-1
Original file line numberDiff line numberDiff line change
@@ -171,7 +171,9 @@
171171

172172
for subset in ["small", "imbalanced", "plus"]:
173173
card = TaskCard(
174-
loader=LoadHF(path="clinc_oos", name=subset),
174+
loader=LoadHF(
175+
path="clinc_oos", name=subset, all_splits=["train", "validation", "test"]
176+
),
175177
preprocess_steps=[
176178
Shuffle(page_size=sys.maxsize),
177179
Rename(field_to_field={"intent": "label"}),

‎prepare/cards/cnn_dailymail.py

+3-1
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,9 @@
99
from unitxt.test_utils.card import test_card
1010

1111
card = TaskCard(
12-
loader=LoadHF(path="cnn_dailymail", name="3.0.0"),
12+
loader=LoadHF(
13+
path="cnn_dailymail", name="3.0.0", all_splits=["train", "validation", "test"]
14+
),
1315
preprocess_steps=[
1416
Rename(field_to_field={"article": "document"}),
1517
Wrap(field="highlights", inside="list", to_field="summaries"),

‎prepare/cards/coedit.py

+8-1
Original file line numberDiff line numberDiff line change
@@ -22,6 +22,7 @@
2222
path="grammarly/coedit",
2323
streaming=True,
2424
filtering_lambda="lambda x: x['task'] == 'gec'",
25+
all_splits=["train", "validation"],
2526
),
2627
preprocess_steps=[
2728
"splitters.small_no_test",
@@ -58,6 +59,7 @@
5859
path="grammarly/coedit",
5960
streaming=True,
6061
filtering_lambda="lambda x: x['task'] == 'gec'",
62+
all_splits=["train", "validation"],
6163
),
6264
preprocess_steps=[
6365
"splitters.small_no_test",
@@ -101,7 +103,9 @@
101103

102104

103105
card = TaskCard(
104-
loader=LoadHF(path="grammarly/coedit", streaming=True),
106+
loader=LoadHF(
107+
path="grammarly/coedit", streaming=True, all_splits=["train", "validation"]
108+
),
105109
preprocess_steps=[
106110
Shuffle(page_size=sys.maxsize),
107111
"splitters.small_no_test",
@@ -146,6 +150,7 @@
146150
path="grammarly/coedit",
147151
streaming=True,
148152
filtering_lambda="lambda x: x['task'] in ['gec', 'simplification', 'coherence', 'neutralize']",
153+
all_splits=["train", "validation"],
149154
),
150155
preprocess_steps=[
151156
Shuffle(page_size=sys.maxsize),
@@ -211,6 +216,7 @@
211216
path="grammarly/coedit",
212217
streaming=True,
213218
filtering_lambda="lambda x: x['task'] in ['gec', 'simplification', 'coherence', 'neutralize']",
219+
all_splits=["train", "validation"],
214220
),
215221
preprocess_steps=[
216222
Shuffle(page_size=sys.maxsize),
@@ -272,6 +278,7 @@
272278
path="grammarly/coedit",
273279
streaming=True,
274280
filtering_lambda="lambda x: x['task'] == 'paraphrase'",
281+
all_splits=["train", "validation"],
275282
),
276283
preprocess_steps=[
277284
"splitters.small_no_test",

‎prepare/cards/cohere_for_ai.py

+1
Original file line numberDiff line numberDiff line change
@@ -24,6 +24,7 @@
2424
name=subset,
2525
streaming=True,
2626
filtering_lambda=f'lambda instance: instance["language"]=="{lang}"',
27+
all_splits=["test"],
2728
),
2829
preprocess_steps=[
2930
SplitRandomMix(

‎prepare/cards/cola.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@
99
from unitxt.test_utils.card import test_card
1010

1111
card = TaskCard(
12-
loader=LoadHF(path="glue", name="cola"),
12+
loader=LoadHF(path="glue", name="cola", all_splits=["train", "validation", "test"]),
1313
preprocess_steps=[
1414
"splitters.small_no_test",
1515
MapInstanceValues(mappers={"label": {"0": "unacceptable", "1": "acceptable"}}),

‎prepare/cards/copa.py

+3-1
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,9 @@
1010
from unitxt.test_utils.card import test_card
1111

1212
card = TaskCard(
13-
loader=LoadHF(path="super_glue", name="copa"),
13+
loader=LoadHF(
14+
path="super_glue", name="copa", all_splits=["test", "train", "validation"]
15+
),
1416
preprocess_steps=[
1517
"splitters.small_no_test",
1618
ListFieldValues(fields=["choice1", "choice2"], to_field="choices"),

‎prepare/cards/coqa.py

+2-2
Original file line numberDiff line numberDiff line change
@@ -6,7 +6,7 @@
66
from unitxt.test_utils.card import test_card
77

88
card = TaskCard(
9-
loader=LoadHF(path="stanfordnlp/coqa"),
9+
loader=LoadHF(path="stanfordnlp/coqa", all_splits=["train", "validation"]),
1010
preprocess_steps=[
1111
"splitters.small_no_test",
1212
Set(fields={"context_type": "story"}),
@@ -62,7 +62,7 @@
6262
add_to_catalog(card, "cards.coqa.qa", overwrite=True)
6363

6464
card = TaskCard(
65-
loader=LoadHF(path="stanfordnlp/coqa"),
65+
loader=LoadHF(path="stanfordnlp/coqa", all_splits=["train", "validation"]),
6666
preprocess_steps=[
6767
"splitters.small_no_test",
6868
Set(fields={"context_type": "dialog", "completion_type": "response"}),

‎prepare/cards/dart.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@
1010
from unitxt.test_utils.card import test_card
1111

1212
card = TaskCard(
13-
loader=LoadHF(path="dart"),
13+
loader=LoadHF(path="dart", all_splits=["train", "validation", "test"]),
1414
preprocess_steps=[
1515
"splitters.small_no_test",
1616
SerializeTriples(field_to_field=[["tripleset", "serialized_triples"]]),

‎prepare/cards/dbpedia_14.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -34,7 +34,7 @@
3434
mappers = {str(i): cls for i, cls in enumerate(classes)}
3535

3636
card = TaskCard(
37-
loader=LoadHF(path=f"{dataset_name}"),
37+
loader=LoadHF(path=f"{dataset_name}", all_splits=["train", "test"]),
3838
preprocess_steps=[
3939
Shuffle(page_size=sys.maxsize),
4040
SplitRandomMix(

0 commit comments

Comments
 (0)