Datasets:
kilt_tasks

+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bin.* filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zstandard filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md ADDED Viewed

	@@ -0,0 +1,25 @@

+# Loading the KILT knowledge source and task data
+The original KILT [release](https://github.com/facebookresearch/KILT) only provides question IDs for the TriviaQA task. Using the full dataset requires mapping those back to the TriviaQA questions, which can be done as follows:
+```python
+from datasets import load_dataset
+# Get the pre-processed Wikipedia knowledge source for kild
+kilt_wiki = load_dataset("kilt_wikipedia")
+# Get the KILT task datasets
+kilt_tasks = load_sataset("kilt_tasks")
+# Most tasks in KILT already have all required data, but KILT-TriviaQA
+# only provides the question IDs, not the questions themselves.
+# Thankfully, we can get the original TriviaQA data with:
+trivia_qa = load_dataset('trivia_qa', 'unfiltered.nocontext')
+# The KILT IDs can then be mapped to the TriviaQA questions with:
+triviaqa_map = {}
+for k in ['train', 'validation', 'test']:
+    triviaqa_map = dict([(q_id, i) for i, q_id in enumerate(triviaqa[k]['question_id'])])
+    kilt_tasks[k + '_triviaqa'] = kilt_tasks[k + '_triviaqa'].filter(lambda x: x['id'] in triviaqa_map)
+    kilt_tasks[k + '_triviaqa'] = kilt_tasks[k + '_triviaqa'].map(lambda x: {'input': triviaqa[k][triviaqa_map[x['id']]]['question']})
+```

dataset_infos.json ADDED Viewed

	@@ -0,0 +1 @@

+ {"all_tasks": {"description": "KILT tasks training and evaluation data.\n- [FEVER](https://fever.ai) | Fact Checking | fever\n- [AIDA CoNLL-YAGO](https://www.mpi-inf.mpg.de/departments/databases-and-information-systems/research/ambiverse-nlu/aida/downloads) | Entity Linking | aidayago2\n- [WNED-WIKI](https://github.com/U-Alberta/wned) | Entity Linking | wned\n- [WNED-CWEB](https://github.com/U-Alberta/wned) | Entity Linking | cweb\n- [T-REx](https://hadyelsahar.github.io/t-rex) | Slot Filling | trex\n- [Zero-Shot RE](http://nlp.cs.washington.edu/zeroshot) | Slot Filling | structured_zeroshot\n- [Natural Questions](https://ai.google.com/research/NaturalQuestions) | Open Domain QA | nq\n- [HotpotQA](https://hotpotqa.github.io) | Open Domain QA | hotpotqa\n- [TriviaQA](http://nlp.cs.washington.edu/triviaqa) | Open Domain QA | triviaqa\n- [ELI5](https://facebookresearch.github.io/ELI5/explore.html) | Open Domain QA | eli5\n- [Wizard of Wikipedia](https://parl.ai/projects/wizard_of_wikipedia) | Dialogue | wow\n\nTo finish linking TriviaQA questions to the IDs provided, follow the instructions [here](http://github.com/huggingface/datasets/datasets/kilt_tasks/README.md).\n", "citation": "@inproceedings{fb_kilt,\n author = {Fabio Petroni and\n Aleksandra Piktus and\n Angela Fan and\n Patrick Lewis and\n Majid Yazdani and\n Nicola De Cao and\n James Thorne and\n Yacine Jernite and\n Vassilis Plachouras and\n Tim Rockt\"aschel and\n Sebastian Riedel},\n title = {{KILT:} a {B}enchmark for {K}nowledge {I}ntensive {L}anguage {T}asks},\n journal = {CoRR},\n archivePrefix = {arXiv},\n year = {2020},\n", "homepage": "https://github.com/facebookresearch/KILT", "license": "", "features": {"id": {"dtype": "string", "id": null, "_type": "Value"}, "input": {"dtype": "string", "id": null, "_type": "Value"}, "meta": {"left_context": {"dtype": "string", "id": null, "_type": "Value"}, "mention": {"dtype": "string", "id": null, "_type": "Value"}, "right_context": {"dtype": "string", "id": null, "_type": "Value"}, "partial_evidence": {"feature": {"start_paragraph_id": {"dtype": "int32", "id": null, "_type": "Value"}, "end_paragraph_id": {"dtype": "int32", "id": null, "_type": "Value"}, "title": {"dtype": "string", "id": null, "_type": "Value"}, "section": {"dtype": "string", "id": null, "_type": "Value"}, "wikipedia_id": {"dtype": "string", "id": null, "_type": "Value"}, "meta": {"feature": {"evidence_span": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}, "obj_surface": {"feature": {"text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "sub_surface": {"feature": {"text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "subj_aliases": {"feature": {"text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "template_questions": {"feature": {"text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "output": {"feature": {"answer": {"dtype": "string", "id": null, "_type": "Value"}, "meta": {"score": {"dtype": "int32", "id": null, "_type": "Value"}}, "provenance": {"feature": {"bleu_score": {"dtype": "float32", "id": null, "_type": "Value"}, "start_character": {"dtype": "int32", "id": null, "_type": "Value"}, "start_paragraph_id": {"dtype": "int32", "id": null, "_type": "Value"}, "end_character": {"dtype": "int32", "id": null, "_type": "Value"}, "end_paragraph_id": {"dtype": "int32", "id": null, "_type": "Value"}, "meta": {"fever_page_id": {"dtype": "string", "id": null, "_type": "Value"}, "fever_sentence_id": {"dtype": "int32", "id": null, "_type": "Value"}, "annotation_id": {"dtype": "string", "id": null, "_type": "Value"}, "yes_no_answer": {"dtype": "string", "id": null, "_type": "Value"}, "evidence_span": {"feature": {"text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "section": {"dtype": "string", "id": null, "_type": "Value"}, "title": {"dtype": "string", "id": null, "_type": "Value"}, "wikipedia_id": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": {"features": null, "resources_checksums": {"train_fever": {}, "validation_fever": {}, "test_fever": {}, "train_nq": {}, "validation_nq": {}, "test_nq": {}, "train_hotpotqa": {}, "validation_hotpotqa": {}, "test_hotpotqa": {}, "train_triviaqa": {}, "validation_triviaqa": {}, "test_triviaqa": {}, "train_eli5": {}, "validation_eli5": {}, "test_eli5": {}, "train_trex": {}, "validation_trex": {}, "test_trex": {}, "train_structured_zeroshot": {}, "validation_structured_zeroshot": {}, "test_structured_zeroshot": {}, "train_aidayago2": {}, "validation_aidayago2": {}, "test_aidayago2": {}, "validation_wned": {}, "test_wned": {}, "validation_cweb": {}, "test_cweb": {}, "train_wow": {}, "validation_wow": {}, "test_wow": {}}}, "supervised_keys": null, "builder_name": "kilt_tasks", "config_name": "all_tasks", "version": {"version_str": "1.0.0", "description": "KILT tasks training and evaluation data", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train_fever": {"name": "train_fever", "num_bytes": 27423152, "num_examples": 104966, "dataset_name": "kilt_tasks"}, "validation_fever": {"name": "validation_fever", "num_bytes": 3945700, "num_examples": 10444, "dataset_name": "kilt_tasks"}, "test_fever": {"name": "test_fever", "num_bytes": 1357227, "num_examples": 10100, "dataset_name": "kilt_tasks"}, "train_aidayago2": {"name": "train_aidayago2", "num_bytes": 69671936, "num_examples": 18395, "dataset_name": "kilt_tasks"}, "validation_aidayago2": {"name": "validation_aidayago2", "num_bytes": 20932848, "num_examples": 4784, "dataset_name": "kilt_tasks"}, "test_aidayago2": {"name": "test_aidayago2", "num_bytes": 14350869, "num_examples": 4463, "dataset_name": "kilt_tasks"}, "validation_wned": {"name": "validation_wned", "num_bytes": 12794390, "num_examples": 3396, "dataset_name": "kilt_tasks"}, "test_wned": {"name": "test_wned", "num_bytes": 13187270, "num_examples": 3376, "dataset_name": "kilt_tasks"}, "validation_cweb": {"name": "validation_cweb", "num_bytes": 90041126, "num_examples": 5599, "dataset_name": "kilt_tasks"}, "test_cweb": {"name": "test_cweb", "num_bytes": 99382290, "num_examples": 5543, "dataset_name": "kilt_tasks"}, "train_trex": {"name": "train_trex", "num_bytes": 1213971546, "num_examples": 2284168, "dataset_name": "kilt_tasks"}, "validation_trex": {"name": "validation_trex", "num_bytes": 2630523, "num_examples": 5000, "dataset_name": "kilt_tasks"}, "test_trex": {"name": "test_trex", "num_bytes": 915477, "num_examples": 5000, "dataset_name": "kilt_tasks"}, "train_structured_zeroshot": {"name": "train_structured_zeroshot", "num_bytes": 55988052, "num_examples": 147909, "dataset_name": "kilt_tasks"}, "validation_structured_zeroshot": {"name": "validation_structured_zeroshot", "num_bytes": 1848289, "num_examples": 3724, "dataset_name": "kilt_tasks"}, "test_structured_zeroshot": {"name": "test_structured_zeroshot", "num_bytes": 1297614, "num_examples": 4966, "dataset_name": "kilt_tasks"}, "train_nq": {"name": "train_nq", "num_bytes": 35533981, "num_examples": 87372, "dataset_name": "kilt_tasks"}, "validation_nq": {"name": "validation_nq", "num_bytes": 6587478, "num_examples": 2837, "dataset_name": "kilt_tasks"}, "test_nq": {"name": "test_nq", "num_bytes": 379518, "num_examples": 1444, "dataset_name": "kilt_tasks"}, "train_hotpotqa": {"name": "train_hotpotqa", "num_bytes": 34838943, "num_examples": 88869, "dataset_name": "kilt_tasks"}, "validation_hotpotqa": {"name": "validation_hotpotqa", "num_bytes": 2428954, "num_examples": 5600, "dataset_name": "kilt_tasks"}, "test_hotpotqa": {"name": "test_hotpotqa", "num_bytes": 1061929, "num_examples": 5569, "dataset_name": "kilt_tasks"}, "train_triviaqa": {"name": "train_triviaqa", "num_bytes": 62374231, "num_examples": 61844, "dataset_name": "kilt_tasks"}, "validation_triviaqa": {"name": "validation_triviaqa", "num_bytes": 5962201, "num_examples": 5359, "dataset_name": "kilt_tasks"}, "test_triviaqa": {"name": "test_triviaqa", "num_bytes": 547068, "num_examples": 6586, "dataset_name": "kilt_tasks"}, "train_eli5": {"name": "train_eli5", "num_bytes": 531158920, "num_examples": 272634, "dataset_name": "kilt_tasks"}, "validation_eli5": {"name": "validation_eli5", "num_bytes": 13833241, "num_examples": 1507, "dataset_name": "kilt_tasks"}, "test_eli5": {"name": "test_eli5", "num_bytes": 126907, "num_examples": 600, "dataset_name": "kilt_tasks"}, "train_wow": {"name": "train_wow", "num_bytes": 64964362, "num_examples": 94577, "dataset_name": "kilt_tasks"}, "validation_wow": {"name": "validation_wow", "num_bytes": 2142799, "num_examples": 3058, "dataset_name": "kilt_tasks"}, "test_wow": {"name": "test_wow", "num_bytes": 1432546, "num_examples": 2944, "dataset_name": "kilt_tasks"}}, "download_checksums": {"http://dl.fbaipublicfiles.com/KILT/fever-train-kilt.jsonl": {"num_bytes": 38941824, "checksum": "a42b948957680b3d316bbc9c24f3e499f0e93a3b0a8b94ca9d972d5da5758c6a"}, "http://dl.fbaipublicfiles.com/KILT/fever-dev-kilt.jsonl": {"num_bytes": 6174139, "checksum": "0132fb971e4206c8bd9da3916f0f46a30ee6610394aee1738ce6cf6644592739"}, "http://dl.fbaipublicfiles.com/KILT/fever-test_without_answers-kilt.jsonl": {"num_bytes": 838585, "checksum": "d95e0a7086c1d5687984460aa9d5b82b3afb58972132c8cc04a75dfd23cb1a86"}, "http://dl.fbaipublicfiles.com/KILT/aidayago2-train-kilt.jsonl": {"num_bytes": 70139831, "checksum": "398231ec4ee9fd5616456d049875e51b1e0709f00ac2d74398dd4480ae5647b2"}, "http://dl.fbaipublicfiles.com/KILT/aidayago2-dev-kilt.jsonl": {"num_bytes": 21061554, "checksum": "3f5e23e1af46c8671c9870e13ced13fd517029da4963ad9ff4b834c4b42deb52"}, "http://dl.fbaipublicfiles.com/KILT/aidayago2-test_without_answers-kilt.jsonl": {"num_bytes": 14436143, "checksum": "2ba0836d3f8c70126022a83fd78da0bf7c910fa7813996dd2afee5e38de63648"}, "http://dl.fbaipublicfiles.com/KILT/wned-dev-kilt.jsonl": {"num_bytes": 12868348, "checksum": "e8e91d120abcb3433c952541aebd8dbf0c6abe3c378bdd64cd6d2186738cfbaf"}, "http://dl.fbaipublicfiles.com/KILT/wned-test_without_answers-kilt.jsonl": {"num_bytes": 13295124, "checksum": "f7b220d404d474f617b00c4a030a87024835b72aff131016716b84feca4d95ee"}, "http://dl.fbaipublicfiles.com/KILT/cweb-dev-kilt.jsonl": {"num_bytes": 90228527, "checksum": "c98b2301818b53df2025b09801be095152244dbbe7242efbfffd018929bfb4bf"}, "http://dl.fbaipublicfiles.com/KILT/cweb-test_without_answers-kilt.jsonl": {"num_bytes": 100216209, "checksum": "18b078340bde25e42c7136c5bc80c2e3557803e40137b40230af5834e74b6c0d"}, "http://dl.fbaipublicfiles.com/KILT/trex-train-kilt.jsonl": {"num_bytes": 1752330104, "checksum": "724a8a0fd1f99fe9b15cf6309d7ce63dcb7cd31f33aea717ff26e6d6b6fbdd97"}, "http://dl.fbaipublicfiles.com/KILT/trex-dev-kilt.jsonl": {"num_bytes": 3803558, "checksum": "7ee1a2cd351d928d01ede1763ae8d6ae442a00b07deef80e347852f20a029e5f"}, "http://dl.fbaipublicfiles.com/KILT/trex-test_without_answers-kilt.jsonl": {"num_bytes": 895854, "checksum": "0a2c6a31ebea567e7a8b31c44bf2fb59dab601c8da6de7f3a53419c7710bdc43"}, "http://dl.fbaipublicfiles.com/KILT/structured_zeroshot-train-kilt.jsonl": {"num_bytes": 71444475, "checksum": "f666359fa2b23e75d700bd1f25efd12d3bf981b1be70d8ad2d268d67e9bb3a5d"}, "http://dl.fbaipublicfiles.com/KILT/structured_zeroshot-dev-kilt.jsonl": {"num_bytes": 2266707, "checksum": "dd120ba5b62c6499f4e79bf5bf5a908eb96b583ba9bf8c76a4535ac21c5dac8d"}, "http://dl.fbaipublicfiles.com/KILT/structured_zeroshot-test_without_answers-kilt.jsonl": {"num_bytes": 1216038, "checksum": "3353137527c6fe9e48910e2c4d94d3ac01352c52e90a618beaaa807968fd39da"}, "http://dl.fbaipublicfiles.com/KILT/nq-train-kilt.jsonl": {"num_bytes": 51895886, "checksum": "13629ef9f0c4bb7b7b237ce08fae2334acb32a2c1bf69db2350021cad46188a4"}, "http://dl.fbaipublicfiles.com/KILT/nq-dev-kilt.jsonl": {"num_bytes": 7936566, "checksum": "1bcdb150fb10504bd80b915730cde1b384c21647bcf14a4592e4e5df42cc6dfb"}, "http://dl.fbaipublicfiles.com/KILT/nq-test_without_answers-kilt.jsonl": {"num_bytes": 334047, "checksum": "1232f95e7d008669930c441444aa470d54678d886335175e2227455716309e15"}, "http://dl.fbaipublicfiles.com/KILT/hotpotqa-train-kilt.jsonl": {"num_bytes": 52767068, "checksum": "7ba73ede642336703d1235c63ecb4c20ae90c7eb19d36302bffa7eb1ec7f9153"}, "http://dl.fbaipublicfiles.com/KILT/hotpotqa-dev-kilt.jsonl": {"num_bytes": 3971321, "checksum": "7f2efbc22c462eb5710e5ecd99447702e7f48203eecd4489bf32a332389d70b6"}, "http://dl.fbaipublicfiles.com/KILT/hotpotqa-test_without_answers-kilt.jsonl": {"num_bytes": 778249, "checksum": "656d51f53796e3ea2a1f808a22160e7363a24a05f1ac7dd094d25659d963ef51"}, "http://dl.fbaipublicfiles.com/KILT/triviaqa-train_id-kilt.jsonl": {"num_bytes": 101613288, "checksum": "509e627a5e606d3b8018c04cd99415d7dca90d7b9a9f491ba0cf2db8c7e999ad"}, "http://dl.fbaipublicfiles.com/KILT/triviaqa-dev_id-kilt.jsonl": {"num_bytes": 9809706, "checksum": "bb505d8437100589de72256bf0d106a4a72da01df5a977fbc778688b8621d5a6"}, "http://dl.fbaipublicfiles.com/KILT/triviaqa-test_id_without_answers-kilt.jsonl": {"num_bytes": 123354, "checksum": "a6904eeb81a269cc675152c72b31c5b3361b40bf277a2973f2836670c7de4a19"}, "http://dl.fbaipublicfiles.com/KILT/eli5-train-kilt.jsonl": {"num_bytes": 548249898, "checksum": "66f19183bda5f5185fe98a5fe84925f6802d9b53a5bb12728ded760fd80ef0a0"}, "http://dl.fbaipublicfiles.com/KILT/eli5-dev-kilt.jsonl": {"num_bytes": 14149811, "checksum": "10387a08e3277d3e5722305cef8e739279c83de0218e7b3db76b301949822303"}, "http://dl.fbaipublicfiles.com/KILT/eli5-test_without_answers-kilt.jsonl": {"num_bytes": 98951, "checksum": "43f82439f15f3141d115c06fb0dc0697a2a6313705dbdec6668eaefe9925a2c3"}, "http://dl.fbaipublicfiles.com/KILT/wow-train-kilt.jsonl": {"num_bytes": 71861702, "checksum": "91e04200a31daadbc9178382b724dac9be88ea9e76cd7a668e7a9430fe67713a"}, "http://dl.fbaipublicfiles.com/KILT/wow-dev-kilt.jsonl": {"num_bytes": 2418241, "checksum": "3202148ecaf41e77d44a2dec2453452c1ff8b3d646727e2c55ae6d9024e8e6b6"}, "http://dl.fbaipublicfiles.com/KILT/wow-test_without_answers-kilt.jsonl": {"num_bytes": 1292018, "checksum": "f3303ec5d7def4cf178552a8ed17bd01cb23bfeccc3abde147bebfa4bade6d8b"}}, "download_size": 3067457126, "post_processing_size": 0, "dataset_size": 2393111387, "size_in_bytes": 5460568513}}

dummy/all_tasks/1.0.0/dummy_data.zip ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:8b4bfcd2a9dd13fca2f9d53da29271ff8346de131e554080bb57d753bc963a02
+size 8382

kilt_tasks.py ADDED Viewed

	@@ -0,0 +1,291 @@

+# coding=utf-8
+# Copyright 2020 The TensorFlow Datasets Authors and the HuggingFace Datasets Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# Lint as: python3
+"""KILT tasks training and evaluation data"""
+from __future__ import absolute_import, division, print_function
+import json
+import logging
+import datasets
+_CITATION = """\
+@inproceedings{fb_kilt,
+    author    = {Fabio Petroni and
+                 Aleksandra Piktus and
+                 Angela Fan and
+                 Patrick Lewis and
+                 Majid Yazdani and
+                 Nicola De Cao and
+                 James Thorne and
+                 Yacine Jernite and
+                 Vassilis Plachouras and
+                 Tim Rockt\"aschel and
+                 Sebastian Riedel},
+    title     = {{KILT:} a {B}enchmark for {K}nowledge {I}ntensive {L}anguage {T}asks},
+    journal   = {CoRR},
+    archivePrefix = {arXiv},
+    year      = {2020},
+"""
+_DESCRIPTION = """\
+KILT tasks training and evaluation data.
+- [FEVER](https://fever.ai) | Fact Checking | fever
+- [AIDA CoNLL-YAGO](https://www.mpi-inf.mpg.de/departments/databases-and-information-systems/research/ambiverse-nlu/aida/downloads) | Entity Linking | aidayago2
+- [WNED-WIKI](https://github.com/U-Alberta/wned) | Entity Linking | wned
+- [WNED-CWEB](https://github.com/U-Alberta/wned) | Entity Linking | cweb
+- [T-REx](https://hadyelsahar.github.io/t-rex) | Slot Filling | trex
+- [Zero-Shot RE](http://nlp.cs.washington.edu/zeroshot) | Slot Filling | structured_zeroshot
+- [Natural Questions](https://ai.google.com/research/NaturalQuestions) | Open Domain QA  | nq
+- [HotpotQA](https://hotpotqa.github.io) | Open Domain QA | hotpotqa
+- [TriviaQA](http://nlp.cs.washington.edu/triviaqa) | Open Domain QA | triviaqa
+- [ELI5](https://facebookresearch.github.io/ELI5/explore.html) | Open Domain QA | eli5
+- [Wizard of Wikipedia](https://parl.ai/projects/wizard_of_wikipedia) | Dialogue | wow
+To finish linking TriviaQA questions to the IDs provided, follow the instructions [here](http://github.com/huggingface/datasets/datasets/kilt_tasks/README.md).
+"""
+_DATA_URLS = {
+    "fever": {
+        "train": "http://dl.fbaipublicfiles.com/KILT/fever-train-kilt.jsonl",
+        "validation": "http://dl.fbaipublicfiles.com/KILT/fever-dev-kilt.jsonl",
+        "test": "http://dl.fbaipublicfiles.com/KILT/fever-test_without_answers-kilt.jsonl",
+    },
+    "aidayago2": {
+        "train": "http://dl.fbaipublicfiles.com/KILT/aidayago2-train-kilt.jsonl",
+        "validation": "http://dl.fbaipublicfiles.com/KILT/aidayago2-dev-kilt.jsonl",
+        "test": "http://dl.fbaipublicfiles.com/KILT/aidayago2-test_without_answers-kilt.jsonl",
+    },
+    "wned": {
+        "validation": "http://dl.fbaipublicfiles.com/KILT/wned-dev-kilt.jsonl",
+        "test": "http://dl.fbaipublicfiles.com/KILT/wned-test_without_answers-kilt.jsonl",
+    },
+    "cweb": {
+        "validation": "http://dl.fbaipublicfiles.com/KILT/cweb-dev-kilt.jsonl",
+        "test": "http://dl.fbaipublicfiles.com/KILT/cweb-test_without_answers-kilt.jsonl",
+    },
+    "trex": {
+        "train": "http://dl.fbaipublicfiles.com/KILT/trex-train-kilt.jsonl",
+        "validation": "http://dl.fbaipublicfiles.com/KILT/trex-dev-kilt.jsonl",
+        "test": "http://dl.fbaipublicfiles.com/KILT/trex-test_without_answers-kilt.jsonl",
+    },
+    "structured_zeroshot": {
+        "train": "http://dl.fbaipublicfiles.com/KILT/structured_zeroshot-train-kilt.jsonl",
+        "validation": "http://dl.fbaipublicfiles.com/KILT/structured_zeroshot-dev-kilt.jsonl",
+        "test": "http://dl.fbaipublicfiles.com/KILT/structured_zeroshot-test_without_answers-kilt.jsonl",
+    },
+    "nq": {
+        "train": "http://dl.fbaipublicfiles.com/KILT/nq-train-kilt.jsonl",
+        "validation": "http://dl.fbaipublicfiles.com/KILT/nq-dev-kilt.jsonl",
+        "test": "http://dl.fbaipublicfiles.com/KILT/nq-test_without_answers-kilt.jsonl",
+    },
+    "hotpotqa": {
+        "train": "http://dl.fbaipublicfiles.com/KILT/hotpotqa-train-kilt.jsonl",
+        "validation": "http://dl.fbaipublicfiles.com/KILT/hotpotqa-dev-kilt.jsonl",
+        "test": "http://dl.fbaipublicfiles.com/KILT/hotpotqa-test_without_answers-kilt.jsonl",
+    },
+    "triviaqa": {
+        "train": "http://dl.fbaipublicfiles.com/KILT/triviaqa-train_id-kilt.jsonl",
+        "validation": "http://dl.fbaipublicfiles.com/KILT/triviaqa-dev_id-kilt.jsonl",
+        "test": "http://dl.fbaipublicfiles.com/KILT/triviaqa-test_id_without_answers-kilt.jsonl",
+    },
+    "eli5": {
+        "train": "http://dl.fbaipublicfiles.com/KILT/eli5-train-kilt.jsonl",
+        "validation": "http://dl.fbaipublicfiles.com/KILT/eli5-dev-kilt.jsonl",
+        "test": "http://dl.fbaipublicfiles.com/KILT/eli5-test_without_answers-kilt.jsonl",
+    },
+    "wow": {
+        "train": "http://dl.fbaipublicfiles.com/KILT/wow-train-kilt.jsonl",
+        "validation": "http://dl.fbaipublicfiles.com/KILT/wow-dev-kilt.jsonl",
+        "test": "http://dl.fbaipublicfiles.com/KILT/wow-test_without_answers-kilt.jsonl",
+    },
+}
+class KILTTasksConfig(datasets.BuilderConfig):
+    """BuilderConfig for KILTTasks."""
+    def __init__(self, **kwargs):
+        """BuilderConfig for KILTTasks.
+            Args:
+        .
+              **kwargs: keyword arguments forwarded to super.
+        """
+        super(KILTTasksConfig, self).__init__(
+            version=datasets.Version("1.0.0", "KILT tasks training and evaluation data"), **kwargs
+        )
+class KILTTasks(datasets.GeneratorBasedBuilder):
+    """WikipediaKILT: Wikipedia pre-processed for KILT. Version 1.0."""
+    BUILDER_CONFIGS = [
+        KILTTasksConfig(
+            name="all_tasks",
+            description="All KILT tasks traiing and evaluation data",
+        ),
+    ]
+    def _info(self):
+        return datasets.DatasetInfo(
+            description=_DESCRIPTION,
+            features=datasets.Features(
+                {
+                    "id": datasets.Value("string"),
+                    "input": datasets.Value("string"),
+                    "meta": datasets.Features(
+                        {
+                            "left_context": datasets.Value("string"),
+                            "mention": datasets.Value("string"),
+                            "right_context": datasets.Value("string"),
+                            "partial_evidence": datasets.features.Sequence(
+                                {
+                                    "start_paragraph_id": datasets.Value("int32"),
+                                    "end_paragraph_id": datasets.Value("int32"),
+                                    "title": datasets.Value("string"),
+                                    "section": datasets.Value("string"),
+                                    "wikipedia_id": datasets.Value("string"),
+                                    "meta": datasets.features.Sequence(
+                                        {
+                                            "evidence_span": datasets.Value("string"),
+                                        }
+                                    ),
+                                }
+                            ),
+                            "obj_surface": datasets.features.Sequence({"text": datasets.Value("string")}),
+                            "sub_surface": datasets.features.Sequence({"text": datasets.Value("string")}),
+                            "subj_aliases": datasets.features.Sequence({"text": datasets.Value("string")}),
+                            "template_questions": datasets.features.Sequence({"text": datasets.Value("string")}),
+                        }
+                    ),
+                    "output": datasets.features.Sequence(
+                        {
+                            "answer": datasets.Value("string"),
+                            "meta": datasets.Features({"score": datasets.Value("int32")}),
+                            "provenance": datasets.features.Sequence(
+                                {
+                                    "bleu_score": datasets.Value("float32"),
+                                    "start_character": datasets.Value("int32"),
+                                    "start_paragraph_id": datasets.Value("int32"),
+                                    "end_character": datasets.Value("int32"),
+                                    "end_paragraph_id": datasets.Value("int32"),
+                                    "meta": datasets.Features(
+                                        {
+                                            "fever_page_id": datasets.Value("string"),
+                                            "fever_sentence_id": datasets.Value("int32"),
+                                            "annotation_id": datasets.Value("string"),  # int runs into overflow issues
+                                            "yes_no_answer": datasets.Value("string"),
+                                            "evidence_span": datasets.features.Sequence(
+                                                {"text": datasets.Value("string")}
+                                            ),
+                                        }
+                                    ),
+                                    "section": datasets.Value("string"),
+                                    "title": datasets.Value("string"),
+                                    "wikipedia_id": datasets.Value("string"),
+                                }
+                            ),
+                        }
+                    ),
+                }
+            ),
+            # No default supervised_keys (as we have to pass both premise
+            # and hypothesis as input).
+            supervised_keys=None,
+            homepage="https://github.com/facebookresearch/KILT",
+            citation=_CITATION,
+        )
+    def _split_generators(self, dl_manager):
+        file_paths = {}
+        for task_name, task_urls in _DATA_URLS.items():
+            file_paths[task_name] = dl_manager.download_and_extract(task_urls)
+        return [
+            datasets.SplitGenerator(name=split + "_" + task, gen_kwargs={"filepath": downloaded_path})
+            for task, split_paths in file_paths.items()
+            for split, downloaded_path in split_paths.items()
+        ]
+    def _generate_examples(self, filepath):
+        """Generate Wikipedia articles for KILT.
+        Args:
+          filepath: a string
+        Yields:
+          dictionaries representing article data and metadata
+        """
+        logging.info("generating examples from = %s", filepath)
+        with open(filepath, encoding="utf-8") as f:
+            for idx, line in enumerate(f):
+                article = json.loads(line.strip())
+                article["input"] = article.get("input", "")
+                # meta
+                article["meta"] = article.get("meta", {})
+                for k in ["left_context", "mention", "right_context"]:
+                    article["meta"][k] = article["meta"].get(k, "")
+                for k in ["obj_surface", "sub_surface", "subj_aliases", "template_questions"]:
+                    article["meta"][k] = {"text": article["meta"].get(k, [])}
+                article["meta"]["partial_evidence"] = article["meta"].get("partial_evidence", [])
+                if "partial_evidence" in article["meta"]:
+                    dct_list = {}
+                    for k in ["start_paragraph_id", "end_paragraph_id"]:
+                        dct_list[k] = [dct.get(k, -1) for dct in article["meta"]["partial_evidence"]]
+                    for k in ["title", "section", "wikipedia_id"]:
+                        dct_list[k] = [dct.get(k, "") for dct in article["meta"]["partial_evidence"]]
+                    if any(["meta" in dct for dct in article["meta"]["partial_evidence"]]):
+                        dct_list["meta"] = [dct.get("meta", {}) for dct in article["meta"]["partial_evidence"]]
+                        for meta in dct_list["meta"]:
+                            meta["evidence_span"] = meta.get("evidence_span", [])
+                    else:
+                        dct_list["meta"] = []
+                    article["meta"]["partial_evidence"] = dct_list
+                # output
+                article["output"] = article.get("output", [])
+                dct_list = {}
+                dct_list["answer"] = [dct.get("answer", "") for dct in article["output"]]
+                if any(["meta" in dct for dct in article["output"]]):
+                    dct_list["meta"] = [dct.get("meta", {"score": 0}) for dct in article["output"]]
+                else:
+                    dct_list["meta"] = []
+                dct_list["provenance"] = []
+                for dct in article["output"]:
+                    if "provenance" in dct:
+                        prov_list = dct["provenance"]
+                        prov_dct_list = {}
+                        prov_dct_list["bleu_score"] = [prov.get("bleu_score", 0.0) for prov in prov_list]
+                        if any(["meta" in prov for prov in prov_list]):
+                            prov_dct_list["meta"] = [prov.get("meta", {}) for prov in prov_list]
+                            for meta_dct in prov_dct_list["meta"]:
+                                meta_dct["fever_page_id"] = meta_dct.get("fever_page_id", "")
+                                meta_dct["fever_sentence_id"] = meta_dct.get("fever_sentence_id", -1)
+                                meta_dct["yes_no_answer"] = meta_dct.get("yes_no_answer", "")
+                                meta_dct["annotation_id"] = str(meta_dct.get("annotation_id", -1))
+                                meta_dct["evidence_span"] = {"text": meta_dct.get("evidence_span", [])}
+                        else:
+                            prov_dct_list["meta"] = []
+                        for k in ["start_character", "start_paragraph_id", "end_character", "end_paragraph_id"]:
+                            prov_dct_list[k] = [prov.get(k, -1) for prov in prov_list]
+                        for k in ["section", "title", "wikipedia_id"]:
+                            prov_dct_list[k] = [prov.get(k, "") for prov in prov_list]
+                        dct_list["provenance"] += [prov_dct_list]
+                article["output"] = dct_list
+                yield idx, article