replace download_and_extract by simple download when there's no extraction in datasets scripts [A-D]

lhoestq · lhoestq · commit 97bce1cfdc97 · 2021-04-28T19:35:52.000+02:00
diff --git a/datasets/acronym_identification/acronym_identification.py b/datasets/acronym_identification/acronym_identification.py
@@ -59,9 +59,9 @@ def _info(self):
         )
 
     def _split_generators(self, dl_manager):
-        train_path = dl_manager.download_and_extract(_TRAIN_URL)
-        valid_path = dl_manager.download_and_extract(_VALID_URL)
-        test_path = dl_manager.download_and_extract(_TEST_URL)
+        train_path = dl_manager.download(_TRAIN_URL)
+        valid_path = dl_manager.download(_VALID_URL)
+        test_path = dl_manager.download(_TEST_URL)
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
diff --git a/datasets/ade_corpus_v2/ade_corpus_v2.py b/datasets/ade_corpus_v2/ade_corpus_v2.py
@@ -158,9 +158,9 @@ def _info(self):
     def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
 
-        DAE_path = dl_manager.download_and_extract(_DOWNLOAD_URL.format("DRUG", "AE", "rel"))
-        DD_path = dl_manager.download_and_extract(_DOWNLOAD_URL.format("DRUG", "DOSE", "rel"))
-        DAE_NEG_path = dl_manager.download_and_extract(_DOWNLOAD_URL.format("ADE", "NEG", "txt"))
+        DAE_path = dl_manager.download(_DOWNLOAD_URL.format("DRUG", "AE", "rel"))
+        DD_path = dl_manager.download(_DOWNLOAD_URL.format("DRUG", "DOSE", "rel"))
+        DAE_NEG_path = dl_manager.download(_DOWNLOAD_URL.format("ADE", "NEG", "txt"))
 
         return [
             datasets.SplitGenerator(
diff --git a/datasets/ag_news/ag_news.py b/datasets/ag_news/ag_news.py
@@ -69,8 +69,8 @@ def _info(self):
         )
 
     def _split_generators(self, dl_manager):
-        train_path = dl_manager.download_and_extract(_TRAIN_DOWNLOAD_URL)
-        test_path = dl_manager.download_and_extract(_TEST_DOWNLOAD_URL)
+        train_path = dl_manager.download(_TRAIN_DOWNLOAD_URL)
+        test_path = dl_manager.download(_TEST_DOWNLOAD_URL)
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": train_path}),
             datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": test_path}),
diff --git a/datasets/alt/alt.py b/datasets/alt/alt.py
@@ -194,7 +194,7 @@ def _split_generators(self, dl_manager):
         else:
             data_split = {}
             for k in _SPLIT:
-                data_split[k] = dl_manager.download_and_extract(_SPLIT[k])
+                data_split[k] = dl_manager.download(_SPLIT[k])
 
             return [
                 datasets.SplitGenerator(
diff --git a/datasets/amazon_reviews_multi/amazon_reviews_multi.py b/datasets/amazon_reviews_multi/amazon_reviews_multi.py
@@ -112,9 +112,9 @@ def _split_generators(self, dl_manager):
         dev_urls = [_DOWNLOAD_URL.format(split="dev", lang=lang) for lang in self.config.languages]
         test_urls = [_DOWNLOAD_URL.format(split="test", lang=lang) for lang in self.config.languages]
 
-        train_paths = dl_manager.download_and_extract(train_urls)
-        dev_paths = dl_manager.download_and_extract(dev_urls)
-        test_paths = dl_manager.download_and_extract(test_urls)
+        train_paths = dl_manager.download(train_urls)
+        dev_paths = dl_manager.download(dev_urls)
+        test_paths = dl_manager.download(test_urls)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"file_paths": train_paths}),
diff --git a/datasets/amttl/amttl.py b/datasets/amttl/amttl.py
@@ -107,7 +107,7 @@ def _split_generators(self, dl_manager):
             "dev": f"{_URL}{_DEV_FILE}",
             "test": f"{_URL}{_TEST_FILE}",
         }
-        downloaded_files = dl_manager.download_and_extract(urls_to_download)
+        downloaded_files = dl_manager.download(urls_to_download)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),
diff --git a/datasets/app_reviews/app_reviews.py b/datasets/app_reviews/app_reviews.py
@@ -58,7 +58,7 @@ def _info(self):
         )
 
     def _split_generators(self, dl_manager):
-        train_path = dl_manager.download_and_extract(_TRAIN_DOWNLOAD_URL)
+        train_path = dl_manager.download(_TRAIN_DOWNLOAD_URL)
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": train_path}),
         ]
diff --git a/datasets/ar_res_reviews/ar_res_reviews.py b/datasets/ar_res_reviews/ar_res_reviews.py
@@ -67,7 +67,7 @@ def _info(self):
     def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
 
-        data_dir = dl_manager.download_and_extract(_DOWNLOAD_URL)
+        data_dir = dl_manager.download(_DOWNLOAD_URL)
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": data_dir}),
         ]
diff --git a/datasets/arabic_pos_dialect/arabic_pos_dialect.py b/datasets/arabic_pos_dialect/arabic_pos_dialect.py
@@ -102,7 +102,7 @@ def _split_generators(self, dl_manager):
         # dl_manager is a datasets.download.DownloadManager that can be used to
         # download and extract URLs
         urls_to_download = {dialect: _URL + "seg_plus_pos_{}.txt".format(dialect) for dialect in _DIALECTS}
-        dl_dir = dl_manager.download_and_extract(urls_to_download)
+        dl_dir = dl_manager.download(urls_to_download)
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
diff --git a/datasets/arcd/arcd.py b/datasets/arcd/arcd.py
@@ -83,7 +83,7 @@ def _info(self):
 
     def _split_generators(self, dl_manager):
         urls_to_download = _URLs
-        downloaded_files = dl_manager.download_and_extract(urls_to_download)
+        downloaded_files = dl_manager.download(urls_to_download)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),
diff --git a/datasets/asset/asset.py b/datasets/asset/asset.py
@@ -113,7 +113,7 @@ def _info(self):
         )
 
     def _split_generators(self, dl_manager):
-        data_dir = dl_manager.download_and_extract(_URLs)
+        data_dir = dl_manager.download(_URLs)
         if self.config.name == "simplification":
             return [
                 datasets.SplitGenerator(
diff --git a/datasets/banking77/banking77.py b/datasets/banking77/banking77.py
@@ -151,8 +151,8 @@ def _info(self):
 
     def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
-        train_path = dl_manager.download_and_extract(_TRAIN_DOWNLOAD_URL)
-        test_path = dl_manager.download_and_extract(_TEST_DOWNLOAD_URL)
+        train_path = dl_manager.download(_TRAIN_DOWNLOAD_URL)
+        test_path = dl_manager.download(_TEST_DOWNLOAD_URL)
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": train_path}),
             datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": test_path}),
diff --git a/datasets/bbc_hindi_nli/bbc_hindi_nli.py b/datasets/bbc_hindi_nli/bbc_hindi_nli.py
@@ -133,9 +133,9 @@ def _info(self):
 
     def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
-        train_path = dl_manager.download_and_extract(_TRAIN_DOWNLOAD_URL)
-        test_path = dl_manager.download_and_extract(_TEST_DOWNLOAD_URL)
-        valid_path = dl_manager.download_and_extract(_VALID_DOWNLOAD_URL)
+        train_path = dl_manager.download(_TRAIN_DOWNLOAD_URL)
+        test_path = dl_manager.download(_TEST_DOWNLOAD_URL)
+        valid_path = dl_manager.download(_VALID_DOWNLOAD_URL)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": train_path}),
diff --git a/datasets/bc2gm_corpus/bc2gm_corpus.py b/datasets/bc2gm_corpus/bc2gm_corpus.py
@@ -104,7 +104,7 @@ def _split_generators(self, dl_manager):
             "dev": f"{_URL}{_DEV_FILE}",
             "test": f"{_URL}{_TEST_FILE}",
         }
-        downloaded_files = dl_manager.download_and_extract(urls_to_download)
+        downloaded_files = dl_manager.download(urls_to_download)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),
diff --git a/datasets/bing_coronavirus_query_set/bing_coronavirus_query_set.py b/datasets/bing_coronavirus_query_set/bing_coronavirus_query_set.py
@@ -95,7 +95,7 @@ def _base_url(queries_by, start_date, end_date):
             return _BASE_URL.format(queries_by, start_date, end_date)
 
         download_url = _base_url(self.config.queries_by, self.config.start_date, self.config.end_date)
-        path = dl_manager.download_and_extract(download_url)
+        path = dl_manager.download(download_url)
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
diff --git a/datasets/blimp/blimp.py b/datasets/blimp/blimp.py
@@ -166,7 +166,7 @@ def _split_generators(self, dl_manager):
         cfg = self.config
         download_urls = {cfg.name: _DOWNLOAD_URL + f"data/{cfg.name}.jsonl"}
 
-        downloaded_files = dl_manager.download_and_extract(download_urls)
+        downloaded_files = dl_manager.download(download_urls)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files[cfg.name]})
diff --git a/datasets/bn_hate_speech/bn_hate_speech.py b/datasets/bn_hate_speech/bn_hate_speech.py
@@ -73,7 +73,7 @@ def _info(self):
     def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
 
-        train_path = dl_manager.download_and_extract(_URL)
+        train_path = dl_manager.download(_URL)
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": train_path}),
         ]
diff --git a/datasets/bprec/bprec.py b/datasets/bprec/bprec.py
@@ -171,7 +171,7 @@ def _split_generators(self, dl_manager):
         # By default the archives will be extracted and a path to a cached folder where they are extracted is returned instead of the archive
         _my_urls = [_URLs[cat] for cat in self.config.categories]
 
-        downloaded_files = dl_manager.download_and_extract(_my_urls)
+        downloaded_files = dl_manager.download(_my_urls)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filedirs": downloaded_files}),
diff --git a/datasets/bsd_ja_en/bsd_ja_en.py b/datasets/bsd_ja_en/bsd_ja_en.py
@@ -107,7 +107,7 @@ def _info(self):
 
     def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
-        data_dir = dl_manager.download_and_extract(_URLs)
+        data_dir = dl_manager.download(_URLs)
 
         return [
             datasets.SplitGenerator(
diff --git a/datasets/c3/c3.py b/datasets/c3/c3.py
@@ -108,7 +108,7 @@ def _split_generators(self, dl_manager):
         # m or d
         T = self.config.type_[0]
         files = [_URL + f"c3-{T}-{split}.json" for split in ["train", "test", "dev"]]
-        dl_dir = dl_manager.download_and_extract(files)
+        dl_dir = dl_manager.download(files)
 
         return [
             datasets.SplitGenerator(
diff --git a/datasets/circa/circa.py b/datasets/circa/circa.py
@@ -101,7 +101,7 @@ def _info(self):
         )
 
     def _split_generators(self, dl_manager):
-        train_path = dl_manager.download_and_extract(_DATA_URL)
+        train_path = dl_manager.download(_DATA_URL)
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
diff --git a/datasets/clinc_oos/clinc_oos.py b/datasets/clinc_oos/clinc_oos.py
@@ -256,7 +256,7 @@ def _info(self):
         )
 
     def _split_generators(self, dl_manager):
-        file_ = dl_manager.download_and_extract(self.config.data_url)
+        file_ = dl_manager.download(self.config.data_url)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": file_, "split": "train"}),
diff --git a/datasets/cmrc2018/cmrc2018.py b/datasets/cmrc2018/cmrc2018.py
@@ -83,7 +83,7 @@ def _split_generators(self, dl_manager):
         # dl_manager is a datasets.download.DownloadManager that can be used to
         # download and extract URLs
         urls_to_download = {"train": _TRAIN_FILE, "dev": _DEV_FILE, "test": _TEST_FILE}
-        downloaded_files = dl_manager.download_and_extract(urls_to_download)
+        downloaded_files = dl_manager.download(urls_to_download)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),
diff --git a/datasets/coached_conv_pref/coached_conv_pref.py b/datasets/coached_conv_pref/coached_conv_pref.py
@@ -122,7 +122,7 @@ def _info(self):
     def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
 
-        data_dir = dl_manager.download_and_extract(_URLs)
+        data_dir = dl_manager.download(_URLs)
 
         # Dataset is a single corpus (does not contain any split)
         return [
diff --git a/datasets/codah/codah.py b/datasets/codah/codah.py
@@ -118,7 +118,7 @@ def _split_generators(self, dl_manager):
             "dev": base_url + "dev.tsv",
             "test": base_url + "test.tsv",
         }
-        downloaded_files = dl_manager.download_and_extract(_urls)
+        downloaded_files = dl_manager.download(_urls)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"data_file": downloaded_files["train"]}),
diff --git a/datasets/com_qa/com_qa.py b/datasets/com_qa/com_qa.py
@@ -82,7 +82,7 @@ def _split_generators(self, dl_manager):
         # dl_manager is a datasets.download.DownloadManager that can be used to
         # download and extract URLs
         urls_to_download = _URLS
-        dl_dir = dl_manager.download_and_extract(urls_to_download)
+        dl_dir = dl_manager.download(urls_to_download)
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
diff --git a/datasets/commonsense_qa/commonsense_qa.py b/datasets/commonsense_qa/commonsense_qa.py
@@ -71,7 +71,7 @@ def _split_generators(self, dl_manager):
 
         download_urls = _URLS
 
-        downloaded_files = dl_manager.download_and_extract(download_urls)
+        downloaded_files = dl_manager.download(download_urls)
 
         return [
             datasets.SplitGenerator(
diff --git a/datasets/conll2000/conll2000.py b/datasets/conll2000/conll2000.py
@@ -171,7 +171,7 @@ def _split_generators(self, dl_manager):
             "train": f"{_URL}{_TRAINING_FILE}",
             "test": f"{_URL}{_TEST_FILE}",
         }
-        downloaded_files = dl_manager.download_and_extract(urls_to_download)
+        downloaded_files = dl_manager.download(urls_to_download)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),
diff --git a/datasets/conll2002/conll2002.py b/datasets/conll2002/conll2002.py
@@ -185,7 +185,7 @@ def _split_generators(self, dl_manager):
             "dev": f"{_URL}{_ES_DEV_FILE if self.config.name == 'es' else _NL_DEV_FILE}",
             "test": f"{_URL}{_ES_TEST_FILE if self.config.name == 'es' else _NL_TEST_FILE}",
         }
-        downloaded_files = dl_manager.download_and_extract(urls_to_download)
+        downloaded_files = dl_manager.download(urls_to_download)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),
diff --git a/datasets/conll2003/conll2003.py b/datasets/conll2003/conll2003.py
@@ -193,7 +193,7 @@ def _split_generators(self, dl_manager):
             "dev": f"{_URL}{_DEV_FILE}",
             "test": f"{_URL}{_TEST_FILE}",
         }
-        downloaded_files = dl_manager.download_and_extract(urls_to_download)
+        downloaded_files = dl_manager.download(urls_to_download)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),
diff --git a/datasets/conllpp/conllpp.py b/datasets/conllpp/conllpp.py
@@ -179,7 +179,7 @@ def _split_generators(self, dl_manager):
             "dev": f"{_URL}{_DEV_FILE}",
             "test": f"{_URL}{_TEST_FILE}",
         }
-        downloaded_files = dl_manager.download_and_extract(urls_to_download)
+        downloaded_files = dl_manager.download(urls_to_download)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),
diff --git a/datasets/cos_e/cos_e.py b/datasets/cos_e/cos_e.py
@@ -56,7 +56,7 @@
 def _download_and_index_cqa(dl_manager, name):
     """Downloads CQA and returns it, indexed by id, for joining with Cos-E."""
 
-    downloaded_files = dl_manager.download_and_extract(
+    downloaded_files = dl_manager.download(
         {
             "cqa_train": _CQA_V1_11_URL_TRAIN if name == "v1.11" else _CQA_V1_0_URL_TRAIN,
             "cqa_dev": _CQA_V1_11_URL_DEV if name == "v1.11" else _CQA_V1_0_URL_DEV,
@@ -147,15 +147,15 @@ def _split_generators(self, dl_manager):
         cqa_indexed = _download_and_index_cqa(dl_manager, self.config.name)
 
         if self.config.name == "v1.11":
-            files = dl_manager.download_and_extract(
+            files = dl_manager.download(
                 {
                     "dev": [_COS_E_URL + "v1.11/cose_dev_v1.11_processed.jsonl"],
                     "train": [_COS_E_URL + "v1.11/cose_train_v1.11_processed.jsonl"],
                 }
             )
 
         elif self.config.name == "v1.0":
-            files = dl_manager.download_and_extract(
+            files = dl_manager.download(
                 {
                     "dev": [_COS_E_URL + "v1.0/cose_dev_v1.0_processed.jsonl"],
                     "train": [_COS_E_URL + "v1.0/cose_train_v1.0_processed.jsonl"],
diff --git a/datasets/cosmos_qa/cosmos_qa.py b/datasets/cosmos_qa/cosmos_qa.py
@@ -71,7 +71,7 @@ def _split_generators(self, dl_manager):
         # dl_manager is a datasets.download.DownloadManager that can be used to
         # download and extract URLs
         urls_to_download = _URLS
-        dl_dir = dl_manager.download_and_extract(urls_to_download)
+        dl_dir = dl_manager.download(urls_to_download)
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
diff --git a/datasets/covid_qa_castorini/covid_qa_castorini.py b/datasets/covid_qa_castorini/covid_qa_castorini.py
@@ -82,7 +82,7 @@ def _info(self):
 
     def _split_generators(self, dl_manager):
         url = _URLs[self.config.name]
-        downloaded_filepath = dl_manager.download_and_extract(url)
+        downloaded_filepath = dl_manager.download(url)
 
         return [
             datasets.SplitGenerator(
diff --git a/datasets/covid_qa_deepset/covid_qa_deepset.py b/datasets/covid_qa_deepset/covid_qa_deepset.py
@@ -80,7 +80,7 @@ def _info(self):
 
     def _split_generators(self, dl_manager):
         url = _URLs[self.config.name]
-        downloaded_filepath = dl_manager.download_and_extract(url)
+        downloaded_filepath = dl_manager.download(url)
 
         return [
             datasets.SplitGenerator(
diff --git a/datasets/covid_tweets_japanese/covid_tweets_japanese.py b/datasets/covid_tweets_japanese/covid_tweets_japanese.py
@@ -68,7 +68,6 @@ def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
 
         my_urls = _URLs["url"]
-        # data_url = dl_manager.download_and_extract(my_urls)
         data_url = dl_manager.download(my_urls)
 
         return [
diff --git a/datasets/craigslist_bargains/craigslist_bargains.py b/datasets/craigslist_bargains/craigslist_bargains.py
@@ -106,7 +106,7 @@ def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
 
         my_urls = _URLs
-        data_dir = dl_manager.download_and_extract(my_urls)
+        data_dir = dl_manager.download(my_urls)
 
         return [
             datasets.SplitGenerator(
diff --git a/datasets/crawl_domain/crawl_domain.py b/datasets/crawl_domain/crawl_domain.py
@@ -68,9 +68,9 @@ def _info(self):
     def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
 
-        train_path = dl_manager.download_and_extract(_URLs["train"])
-        test_path = dl_manager.download_and_extract(_URLs["test"])
-        dev_path = dl_manager.download_and_extract(_URLs["dev"])
+        train_path = dl_manager.download(_URLs["train"])
+        test_path = dl_manager.download(_URLs["test"])
+        dev_path = dl_manager.download(_URLs["dev"])
 
         return [
             datasets.SplitGenerator(
diff --git a/datasets/crime_and_punish/crime_and_punish.py b/datasets/crime_and_punish/crime_and_punish.py
@@ -59,7 +59,7 @@ def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
 
         if self.config.name == "crime-and-punish":
-            data = dl_manager.download_and_extract(self.config.data_url)
+            data = dl_manager.download(self.config.data_url)
 
             return [
                 datasets.SplitGenerator(
diff --git a/datasets/cs_restaurants/cs_restaurants.py b/datasets/cs_restaurants/cs_restaurants.py
diff --git a/datasets/curiosity_dialogs/curiosity_dialogs.py b/datasets/curiosity_dialogs/curiosity_dialogs.py
diff --git a/datasets/danish_political_comments/danish_political_comments.py b/datasets/danish_political_comments/danish_political_comments.py
diff --git a/datasets/deal_or_no_dialog/deal_or_no_dialog.py b/datasets/deal_or_no_dialog/deal_or_no_dialog.py
diff --git a/datasets/ethos/ethos.py b/datasets/ethos/ethos.py

Original file line number	Diff line number	Diff line change
`@@ -107,7 +107,7 @@ def _split_generators(self, dl_manager):`
`107`	`107`	`"dev": f"{_URL}{_DEV_FILE}",`
`108`	`108`	`"test": f"{_URL}{_TEST_FILE}",`
`109`	`109`	`}`
`110`		`- downloaded_files = dl_manager.download_and_extract(urls_to_download)`
	`110`	`+ downloaded_files = dl_manager.download(urls_to_download)`
`111`	`111`
`112`	`112`	`return [`
`113`	`113`	`datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),`
Original file line number	Diff line number	Diff line change
`@@ -58,7 +58,7 @@ def _info(self):`
`58`	`58`	`)`
`59`	`59`
`60`	`60`	`def _split_generators(self, dl_manager):`
`61`		`- train_path = dl_manager.download_and_extract(_TRAIN_DOWNLOAD_URL)`
	`61`	`+ train_path = dl_manager.download(_TRAIN_DOWNLOAD_URL)`
`62`	`62`	`return [`
`63`	`63`	`datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": train_path}),`
`64`	`64`	`]`
Original file line number	Diff line number	Diff line change
`@@ -67,7 +67,7 @@ def _info(self):`
`67`	`67`	`def _split_generators(self, dl_manager):`
`68`	`68`	`"""Returns SplitGenerators."""`
`69`	`69`
`70`		`- data_dir = dl_manager.download_and_extract(_DOWNLOAD_URL)`
	`70`	`+ data_dir = dl_manager.download(_DOWNLOAD_URL)`
`71`	`71`	`return [`
`72`	`72`	`datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": data_dir}),`
`73`	`73`	`]`