Convert dataset to Parquet (#4)

- Convert dataset to Parquet (e41544f1cce95e1e671a8c93438616372ad49fd4)
- Add ARC-Easy data files (4d4ed1b0a54f45df33cd8eb45c157cc00cdaf35e)
- Delete loading script (a47c900f11d268f8922aa5d672e71e8e2c58f27c)
- Delete legacy dataset_infos.json (0555bd897c85012514721df4e419709132bb00d4)

ARC-Challenge/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62f03257e737aed263f55c6abf87c7bb0028a44a6bdd2a26eb1279eb42c1d1e9
+size 203808

ARC-Challenge/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e488c1587ffdcfc8443f916c53488a95cd471c5790e0746c6bfe4cecf20962cb
+size 189909

ARC-Challenge/validation-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:395a5c88d1580d69855fbaee9450270578df1ad5af6259771cd0a42c20e99f05
+size 55743

ARC-Easy/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4160597d618ae851c7eb04e281574f3f654776216ac6b6641588d64527b47177
+size 346257

ARC-Easy/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b315db8a4be597dc7daa50a4e70d48dd7c990c32085629e6ccd8c926beaa80b5
+size 330598

ARC-Easy/validation-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed890ff1e4cef7a7140d3a30dcea3ed2c9d467c6458f447ad9ef0176d8dcbb74
+size 86080

README.md

@@ -5,8 +5,6 @@ language_creators:
 - found
 language:
 - en
-language_bcp47:
-- en-US
 license:
 - cc-by-sa-4.0
 multilinguality:
@@ -20,8 +18,9 @@ task_categories:
 task_ids:
 - open-domain-qa
 - multiple-choice-qa
-paperswithcode_id: null
 pretty_name: Ai2Arc
+language_bcp47:
+- en-US
 dataset_info:
 - config_name: ARC-Challenge
   features:
@@ -39,16 +38,16 @@ dataset_info:
     dtype: string
   splits:
   - name: train
-    num_bytes: 351888
+    num_bytes: 349760
     num_examples: 1119
   - name: test
-    num_bytes: 377740
+    num_bytes: 375511
     num_examples: 1172
   - name: validation
-    num_bytes: 97254
+    num_bytes: 96660
     num_examples: 299
-  download_size: 680841265
-  dataset_size: 826882
+  download_size: 449460
+  dataset_size: 821931
 - config_name: ARC-Easy
   features:
   - name: id
@@ -65,16 +64,33 @@ dataset_info:
     dtype: string
   splits:
   - name: train
-    num_bytes: 623254
+    num_bytes: 619000
     num_examples: 2251
   - name: test
-    num_bytes: 661997
+    num_bytes: 657514
     num_examples: 2376
   - name: validation
-    num_bytes: 158498
+    num_bytes: 157394
     num_examples: 570
-  download_size: 680841265
-  dataset_size: 1443749
+  download_size: 762935
+  dataset_size: 1433908
+configs:
+- config_name: ARC-Challenge
+  data_files:
+  - split: train
+    path: ARC-Challenge/train-*
+  - split: test
+    path: ARC-Challenge/test-*
+  - split: validation
+    path: ARC-Challenge/validation-*
+- config_name: ARC-Easy
+  data_files:
+  - split: train
+    path: ARC-Easy/train-*
+  - split: test
+    path: ARC-Easy/test-*
+  - split: validation
+    path: ARC-Easy/validation-*
 ---
 
 # Dataset Card for "ai2_arc"

ai2_arc.py

@@ -1,132 +0,0 @@
-"""TODO(arc): Add a description here."""
-
-
-import json
-import os
-
-import datasets
-
-
-# TODO(ai2_arc): BibTeX citation
-_CITATION = """\
-@article{allenai:arc,
-      author    = {Peter Clark  and Isaac Cowhey and Oren Etzioni and Tushar Khot and
-                    Ashish Sabharwal and Carissa Schoenick and Oyvind Tafjord},
-      title     = {Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge},
-      journal   = {arXiv:1803.05457v1},
-      year      = {2018},
-}
-"""
-
-# TODO(ai2_arc):
-_DESCRIPTION = """\
-A new dataset of 7,787 genuine grade-school level, multiple-choice science questions, assembled to encourage research in
- advanced question-answering. The dataset is partitioned into a Challenge Set and an Easy Set, where the former contains
- only questions answered incorrectly by both a retrieval-based algorithm and a word co-occurrence algorithm. We are also
- including a corpus of over 14 million science sentences relevant to the task, and an implementation of three neural baseline models for this dataset. We pose ARC as a challenge to the community.
-"""
-
-_URL = "https://s3-us-west-2.amazonaws.com/ai2-website/data/ARC-V1-Feb2018.zip"
-
-
-class Ai2ArcConfig(datasets.BuilderConfig):
-    """BuilderConfig for Ai2ARC."""
-
-    def __init__(self, **kwargs):
-        """BuilderConfig for Ai2Arc.
-
-        Args:
-          **kwargs: keyword arguments forwarded to super.
-        """
-        super(Ai2ArcConfig, self).__init__(version=datasets.Version("1.0.0", ""), **kwargs)
-
-
-class Ai2Arc(datasets.GeneratorBasedBuilder):
-    """TODO(arc): Short description of my dataset."""
-
-    # TODO(arc): Set up version.
-    VERSION = datasets.Version("1.0.0")
-    BUILDER_CONFIGS = [
-        Ai2ArcConfig(
-            name="ARC-Challenge",
-            description="""\
-          Challenge Set of 2590 “hard” questions (those that both a retrieval and a co-occurrence method fail to answer correctly)
-          """,
-        ),
-        Ai2ArcConfig(
-            name="ARC-Easy",
-            description="""\
-          Easy Set of 5197 questions
-          """,
-        ),
-    ]
-
-    def _info(self):
-        # TODO(ai2_arc): Specifies the datasets.DatasetInfo object
-        return datasets.DatasetInfo(
-            # This is the description that will appear on the datasets page.
-            description=_DESCRIPTION,
-            # datasets.features.FeatureConnectors
-            features=datasets.Features(
-                {
-                    "id": datasets.Value("string"),
-                    "question": datasets.Value("string"),
-                    "choices": datasets.features.Sequence(
-                        {"text": datasets.Value("string"), "label": datasets.Value("string")}
-                    ),
-                    "answerKey": datasets.Value("string")
-                    # These are the features of your dataset like images, labels ...
-                }
-            ),
-            # If there's a common (input, target) tuple from the features,
-            # specify them here. They'll be used if as_supervised=True in
-            # builder.as_dataset.
-            supervised_keys=None,
-            # Homepage of the dataset for documentation
-            homepage="https://allenai.org/data/arc",
-            citation=_CITATION,
-        )
-
-    def _split_generators(self, dl_manager):
-        """Returns SplitGenerators."""
-        # TODO(ai2_arc): Downloads the data and defines the splits
-        # dl_manager is a datasets.download.DownloadManager that can be used to
-        # download and extract URLs
-        dl_dir = dl_manager.download_and_extract(_URL)
-        data_dir = os.path.join(dl_dir, "ARC-V1-Feb2018-2")
-        return [
-            datasets.SplitGenerator(
-                name=datasets.Split.TRAIN,
-                # These kwargs will be passed to _generate_examples
-                gen_kwargs={"filepath": os.path.join(data_dir, self.config.name, self.config.name + "-Train.jsonl")},
-            ),
-            datasets.SplitGenerator(
-                name=datasets.Split.TEST,
-                # These kwargs will be passed to _generate_examples
-                gen_kwargs={"filepath": os.path.join(data_dir, self.config.name, self.config.name + "-Test.jsonl")},
-            ),
-            datasets.SplitGenerator(
-                name=datasets.Split.VALIDATION,
-                # These kwargs will be passed to _generate_examples
-                gen_kwargs={"filepath": os.path.join(data_dir, self.config.name, self.config.name + "-Dev.jsonl")},
-            ),
-        ]
-
-    def _generate_examples(self, filepath):
-        """Yields examples."""
-        # TODO(ai2_arc): Yields (key, example) tuples from the dataset
-        with open(filepath, encoding="utf-8") as f:
-            for row in f:
-                data = json.loads(row)
-                answerkey = data["answerKey"]
-                id_ = data["id"]
-                question = data["question"]["stem"]
-                choices = data["question"]["choices"]
-                text_choices = [choice["text"] for choice in choices]
-                label_choices = [choice["label"] for choice in choices]
-                yield id_, {
-                    "id": id_,
-                    "answerKey": answerkey,
-                    "question": question,
-                    "choices": {"text": text_choices, "label": label_choices},
-                }

dataset_infos.json

@@ -1 +0,0 @@
-{"ARC-Challenge": {"description": "A new dataset of 7,787 genuine grade-school level, multiple-choice science questions, assembled to encourage research in\n advanced question-answering. The dataset is partitioned into a Challenge Set and an Easy Set, where the former contains\n only questions answered incorrectly by both a retrieval-based algorithm and a word co-occurrence algorithm. We are also\n including a corpus of over 14 million science sentences relevant to the task, and an implementation of three neural baseline models for this dataset. We pose ARC as a challenge to the community.\n", "citation": "@article{allenai:arc,\n      author    = {Peter Clark  and Isaac Cowhey and Oren Etzioni and Tushar Khot and\n                    Ashish Sabharwal and Carissa Schoenick and Oyvind Tafjord},\n      title     = {Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge},\n      journal   = {arXiv:1803.05457v1},\n      year      = {2018},\n}\n", "homepage": "https://allenai.org/data/arc", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"text": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "supervised_keys": null, "builder_name": "ai2_arc", "config_name": "ARC-Challenge", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 377740, "num_examples": 1172, "dataset_name": "ai2_arc"}, "train": {"name": "train", "num_bytes": 351888, "num_examples": 1119, "dataset_name": "ai2_arc"}, "validation": {"name": "validation", "num_bytes": 97254, "num_examples": 299, "dataset_name": "ai2_arc"}}, "download_checksums": {"https://s3-us-west-2.amazonaws.com/ai2-website/data/ARC-V1-Feb2018.zip": {"num_bytes": 680841265, "checksum": "6d2d5ab50b2ceec6ba5f79c921be77cf2de712ea25a2b3f4fff3acc101cecfa0"}}, "download_size": 680841265, "dataset_size": 826882, "size_in_bytes": 681668147}, "ARC-Easy": {"description": "A new dataset of 7,787 genuine grade-school level, multiple-choice science questions, assembled to encourage research in\n advanced question-answering. The dataset is partitioned into a Challenge Set and an Easy Set, where the former contains\n only questions answered incorrectly by both a retrieval-based algorithm and a word co-occurrence algorithm. We are also\n including a corpus of over 14 million science sentences relevant to the task, and an implementation of three neural baseline models for this dataset. We pose ARC as a challenge to the community.\n", "citation": "@article{allenai:arc,\n      author    = {Peter Clark  and Isaac Cowhey and Oren Etzioni and Tushar Khot and\n                    Ashish Sabharwal and Carissa Schoenick and Oyvind Tafjord},\n      title     = {Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge},\n      journal   = {arXiv:1803.05457v1},\n      year      = {2018},\n}\n", "homepage": "https://allenai.org/data/arc", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"dtype": "string", "id": null, "_type": "Value"}, "choices": {"feature": {"text": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "answerKey": {"dtype": "string", "id": null, "_type": "Value"}}, "supervised_keys": null, "builder_name": "ai2_arc", "config_name": "ARC-Easy", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"test": {"name": "test", "num_bytes": 661997, "num_examples": 2376, "dataset_name": "ai2_arc"}, "train": {"name": "train", "num_bytes": 623254, "num_examples": 2251, "dataset_name": "ai2_arc"}, "validation": {"name": "validation", "num_bytes": 158498, "num_examples": 570, "dataset_name": "ai2_arc"}}, "download_checksums": {"https://s3-us-west-2.amazonaws.com/ai2-website/data/ARC-V1-Feb2018.zip": {"num_bytes": 680841265, "checksum": "6d2d5ab50b2ceec6ba5f79c921be77cf2de712ea25a2b3f4fff3acc101cecfa0"}}, "download_size": 680841265, "dataset_size": 1443749, "size_in_bytes": 682285014}}