Datasets:
Create new file
Browse files
TRIP.py
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import os
|
| 2 |
+
from zipfile import ZipFile
|
| 3 |
+
|
| 4 |
+
import datasets
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
_CITATION = """\
|
| 8 |
+
@misc{storks2021tiered,
|
| 9 |
+
title={Tiered Reasoning for Intuitive Physics: Toward Verifiable Commonsense Language Understanding},
|
| 10 |
+
author={Shane Storks and Qiaozi Gao and Yichi Zhang and Joyce Chai},
|
| 11 |
+
year={2021},
|
| 12 |
+
booktitle={Findings of the Association for Computational Linguistics: EMNLP 2021},
|
| 13 |
+
location={Punta Cana, Dominican Republic},
|
| 14 |
+
publisher={Association for Computational Linguistics},
|
| 15 |
+
}
|
| 16 |
+
"""
|
| 17 |
+
|
| 18 |
+
_DESCRIPTION = """\
|
| 19 |
+
We introduce Tiered Reasoning for Intuitive Physics (TRIP), a novel commonsense reasoning dataset with dense annotations that enable multi-tiered evaluation of machines’ reasoning process.
|
| 20 |
+
"""
|
| 21 |
+
|
| 22 |
+
_HOMEPAGE = "https://huggingface.co/datasets/sled-umich/TRIP"
|
| 23 |
+
|
| 24 |
+
class TRIP(datasets.GeneratorBasedBuilder):
|
| 25 |
+
"""DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset"""
|
| 26 |
+
|
| 27 |
+
VERSION = datasets.Version("1.0.0")
|
| 28 |
+
|
| 29 |
+
__EOU__ = "__eou__"
|
| 30 |
+
|
| 31 |
+
def _info(self):
|
| 32 |
+
return datasets.DatasetInfo(
|
| 33 |
+
description=_DESCRIPTION,
|
| 34 |
+
features=datasets.Features(
|
| 35 |
+
{
|
| 36 |
+
"story_id": datasets.Value("int32"),
|
| 37 |
+
"worker_id": datasets.Value("string"),
|
| 38 |
+
"type": datasets.Value("string"),
|
| 39 |
+
"idx": datasets.Value("int32"),
|
| 40 |
+
"aug": datasets.Value("bool"),
|
| 41 |
+
"actor": datasets.Value("string"),
|
| 42 |
+
"location": datasets.Value("string"),
|
| 43 |
+
"objects": datasets.Value("string"),
|
| 44 |
+
"sentences": datasets.features.Sequence(datasets.Value("string")),
|
| 45 |
+
"length": datasets.Value("int32"),
|
| 46 |
+
"example_id": datasets.Value("string"),
|
| 47 |
+
"plausible": datasets.Value("bool"),
|
| 48 |
+
"breakpoint": datasets.Value("int32"),
|
| 49 |
+
"confl_sents": datasets.features.Sequence(datasets.Value("int32")),
|
| 50 |
+
"confl_pairs": datasets.features.Sequence({datasets.Value("int32"), datasets.Value("int32")}),
|
| 51 |
+
"confl_states": datasets.features.Sequence(datasets.Value("int32")),
|
| 52 |
+
}
|
| 53 |
+
),
|
| 54 |
+
supervised_keys=None,
|
| 55 |
+
homepage=_HOMEPAGE,
|
| 56 |
+
citation=_CITATION,
|
| 57 |
+
)
|