update dataset

README.md

@@ -12,6 +12,18 @@ configs:
   data_files:
   - path: pubmed_qa/*.arrow
     split: test
+- config_name: squadshifts_amazon
+  data_files:
+  - path: squadshifts_amazon/*.arrow
+    split: test
+- config_name: squadshifts_nyt
+  data_files:
+  - path: squadshifts_nyt/*.arrow
+    split: test
+- config_name: squadshifts_reddit
+  data_files:
+  - path: squadshifts_reddit/*.arrow
+    split: test
 - config_name: vitaminc
   data_files:
   - path: vitaminc/*.arrow

squadshifts_amazon/data-00000-of-00001.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bded198f7681bf2b857f01da6eeb4d5e994b6e7569174b729e6cebef2db8b1d6
+size 9448840

squadshifts_amazon/dataset_info.json

@@ -0,0 +1,63 @@
+{
+  "builder_name": "parquet",
+  "citation": "@InProceedings{pmlr-v119-miller20a,\n  title = {The Effect of Natural Distribution Shift on Question Answering Models},\n  author = {Miller, John and Krauth, Karl and Recht, Benjamin and Schmidt, Ludwig},\n  booktitle = {Proceedings of the 37th International Conference on Machine Learning},\n  pages = {6905--6916},\n  year = {2020},\n  editor = {III, Hal Daum\u00e9 and Singh, Aarti},\n  volume = {119},\n  series = {Proceedings of Machine Learning Research},\n  month = {13--18 Jul},\n  publisher = {PMLR},\n  pdf = {http://proceedings.mlr.press/v119/miller20a/miller20a.pdf},\n  url = {https://proceedings.mlr.press/v119/miller20a.html},\n}\n",
+  "config_name": "amazon",
+  "dataset_name": "squadshifts",
+  "dataset_size": 9443740,
+  "description": "\\\nSquadShifts consists of four new test sets for the Stanford Question Answering \\\nDataset (SQuAD) from four different domains: Wikipedia articles, New York \\\nTimes articles, Reddit comments, and Amazon product reviews. Each dataset \\\nwas generated using the same data generating pipeline, Amazon Mechanical \\\nTurk interface, and data cleaning code as the original SQuAD v1.1 dataset. \\\nThe \"new-wikipedia\" dataset measures overfitting on the original SQuAD v1.1 \\\ndataset.  The \"new-york-times\", \"reddit\", and \"amazon\" datasets measure \\\nrobustness to natural distribution shifts. We encourage SQuAD model developers \\\nto also evaluate their methods on these new datasets! \\\n",
+  "download_checksums": {
+    "hf://datasets/squadshifts@3032f6783da8c01d6e25fcc5b45fccd5eea1c5bb/amazon/test/0000.parquet": {
+      "num_bytes": 1743431,
+      "checksum": null
+    }
+  },
+  "download_size": 1743431,
+  "features": {
+    "id": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "title": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "context": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "question": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "answers": {
+      "feature": {
+        "text": {
+          "dtype": "string",
+          "_type": "Value"
+        },
+        "answer_start": {
+          "dtype": "int32",
+          "_type": "Value"
+        }
+      },
+      "_type": "Sequence"
+    }
+  },
+  "homepage": "https://modestyachts.github.io/squadshifts-website/index.html",
+  "license": "CC-BY-4.0",
+  "size_in_bytes": 11187171,
+  "splits": {
+    "test": {
+      "name": "test",
+      "num_bytes": 9443740,
+      "num_examples": 9885,
+      "dataset_name": "squadshifts"
+    }
+  },
+  "version": {
+    "version_str": "1.0.0",
+    "major": 1,
+    "minor": 0,
+    "patch": 0
+  }
+}

squadshifts_amazon/state.json

@@ -0,0 +1,13 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00001.arrow"
+    }
+  ],
+  "_fingerprint": "0c0f6d1f1d9d0fbc",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": "test"
+}

squadshifts_nyt/data-00000-of-00001.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14c19b776898efb24c0e5b28e661e7b91cda1c9def528cf453930ee0ef267ad7
+size 10796648

squadshifts_nyt/dataset_info.json

@@ -0,0 +1,63 @@
+{
+  "builder_name": "parquet",
+  "citation": "@InProceedings{pmlr-v119-miller20a,\n  title = {The Effect of Natural Distribution Shift on Question Answering Models},\n  author = {Miller, John and Krauth, Karl and Recht, Benjamin and Schmidt, Ludwig},\n  booktitle = {Proceedings of the 37th International Conference on Machine Learning},\n  pages = {6905--6916},\n  year = {2020},\n  editor = {III, Hal Daum\u00e9 and Singh, Aarti},\n  volume = {119},\n  series = {Proceedings of Machine Learning Research},\n  month = {13--18 Jul},\n  publisher = {PMLR},\n  pdf = {http://proceedings.mlr.press/v119/miller20a/miller20a.pdf},\n  url = {https://proceedings.mlr.press/v119/miller20a.html},\n}\n",
+  "config_name": "nyt",
+  "dataset_name": "squadshifts",
+  "dataset_size": 10791235,
+  "description": "\\\nSquadShifts consists of four new test sets for the Stanford Question Answering \\\nDataset (SQuAD) from four different domains: Wikipedia articles, New York \\\nTimes articles, Reddit comments, and Amazon product reviews. Each dataset \\\nwas generated using the same data generating pipeline, Amazon Mechanical \\\nTurk interface, and data cleaning code as the original SQuAD v1.1 dataset. \\\nThe \"new-wikipedia\" dataset measures overfitting on the original SQuAD v1.1 \\\ndataset.  The \"new-york-times\", \"reddit\", and \"amazon\" datasets measure \\\nrobustness to natural distribution shifts. We encourage SQuAD model developers \\\nto also evaluate their methods on these new datasets! \\\n",
+  "download_checksums": {
+    "hf://datasets/squadshifts@3032f6783da8c01d6e25fcc5b45fccd5eea1c5bb/nyt/test/0000.parquet": {
+      "num_bytes": 1949981,
+      "checksum": null
+    }
+  },
+  "download_size": 1949981,
+  "features": {
+    "id": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "title": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "context": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "question": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "answers": {
+      "feature": {
+        "text": {
+          "dtype": "string",
+          "_type": "Value"
+        },
+        "answer_start": {
+          "dtype": "int32",
+          "_type": "Value"
+        }
+      },
+      "_type": "Sequence"
+    }
+  },
+  "homepage": "https://modestyachts.github.io/squadshifts-website/index.html",
+  "license": "CC-BY-4.0",
+  "size_in_bytes": 12741216,
+  "splits": {
+    "test": {
+      "name": "test",
+      "num_bytes": 10791235,
+      "num_examples": 10065,
+      "dataset_name": "squadshifts"
+    }
+  },
+  "version": {
+    "version_str": "1.0.0",
+    "major": 1,
+    "minor": 0,
+    "patch": 0
+  }
+}

squadshifts_nyt/state.json

@@ -0,0 +1,13 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00001.arrow"
+    }
+  ],
+  "_fingerprint": "38a3eaf5f4d060a0",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": "test"
+}

squadshifts_reddit/data-00000-of-00001.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10621890e317f1bfd1a0a845bbcd5cfb2b926315ee093ce3ffb706b71c1f68be
+size 9477504

squadshifts_reddit/dataset_info.json

@@ -0,0 +1,63 @@
+{
+  "builder_name": "parquet",
+  "citation": "@InProceedings{pmlr-v119-miller20a,\n  title = {The Effect of Natural Distribution Shift on Question Answering Models},\n  author = {Miller, John and Krauth, Karl and Recht, Benjamin and Schmidt, Ludwig},\n  booktitle = {Proceedings of the 37th International Conference on Machine Learning},\n  pages = {6905--6916},\n  year = {2020},\n  editor = {III, Hal Daum\u00e9 and Singh, Aarti},\n  volume = {119},\n  series = {Proceedings of Machine Learning Research},\n  month = {13--18 Jul},\n  publisher = {PMLR},\n  pdf = {http://proceedings.mlr.press/v119/miller20a/miller20a.pdf},\n  url = {https://proceedings.mlr.press/v119/miller20a.html},\n}\n",
+  "config_name": "reddit",
+  "dataset_name": "squadshifts",
+  "dataset_size": 9472692,
+  "description": "\\\nSquadShifts consists of four new test sets for the Stanford Question Answering \\\nDataset (SQuAD) from four different domains: Wikipedia articles, New York \\\nTimes articles, Reddit comments, and Amazon product reviews. Each dataset \\\nwas generated using the same data generating pipeline, Amazon Mechanical \\\nTurk interface, and data cleaning code as the original SQuAD v1.1 dataset. \\\nThe \"new-wikipedia\" dataset measures overfitting on the original SQuAD v1.1 \\\ndataset.  The \"new-york-times\", \"reddit\", and \"amazon\" datasets measure \\\nrobustness to natural distribution shifts. We encourage SQuAD model developers \\\nto also evaluate their methods on these new datasets! \\\n",
+  "download_checksums": {
+    "hf://datasets/squadshifts@3032f6783da8c01d6e25fcc5b45fccd5eea1c5bb/reddit/test/0000.parquet": {
+      "num_bytes": 1817051,
+      "checksum": null
+    }
+  },
+  "download_size": 1817051,
+  "features": {
+    "id": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "title": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "context": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "question": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "answers": {
+      "feature": {
+        "text": {
+          "dtype": "string",
+          "_type": "Value"
+        },
+        "answer_start": {
+          "dtype": "int32",
+          "_type": "Value"
+        }
+      },
+      "_type": "Sequence"
+    }
+  },
+  "homepage": "https://modestyachts.github.io/squadshifts-website/index.html",
+  "license": "CC-BY-4.0",
+  "size_in_bytes": 11289743,
+  "splits": {
+    "test": {
+      "name": "test",
+      "num_bytes": 9472692,
+      "num_examples": 9803,
+      "dataset_name": "squadshifts"
+    }
+  },
+  "version": {
+    "version_str": "1.0.0",
+    "major": 1,
+    "minor": 0,
+    "patch": 0
+  }
+}

squadshifts_reddit/state.json

@@ -0,0 +1,13 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00001.arrow"
+    }
+  ],
+  "_fingerprint": "c5d3d87245b1c1e9",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": "test"
+}