Fix CI after killing archive maps (#4724)

* 🐛 Fix model ids for BART and Flaubert

docs/source/model_doc/bart.rst

@@ -22,7 +22,7 @@ Implementation Notes
 - The forward pass of ``BartModel`` will create decoder inputs (using the helper function ``transformers.modeling_bart._prepare_bart_decoder_inputs``)  if they are not passed. This is different than some other modeling APIs.
 - Model predictions are intended to be identical to the original implementation. This only works, however, if the string you pass to ``fairseq.encode`` starts with a space.
 - ``BartForConditionalGeneration.generate`` should be used for conditional generation tasks like summarization, see the example in that docstrings
-- Models that load the ``"bart-large-cnn"`` weights will not have a ``mask_token_id``, or be able to perform mask filling tasks.
+- Models that load the ``"facebook/bart-large-cnn"`` weights will not have a ``mask_token_id``, or be able to perform mask filling tasks.
 
 
 

examples/summarization/bart/evaluate_cnn.py

@@ -21,7 +21,7 @@ def generate_summaries(
 ):
     fout = Path(out_file).open("w")
     model = BartForConditionalGeneration.from_pretrained(model_name).to(device)
-    tokenizer = BartTokenizer.from_pretrained("bart-large")
+    tokenizer = BartTokenizer.from_pretrained("facebook/bart-large")
 
     max_length = 140
     min_length = 55
@@ -54,7 +54,7 @@ def run_generate():
         "output_path", type=str, help="where to save summaries",
     )
     parser.add_argument(
-        "model_name", type=str, default="bart-large-cnn", help="like bart-large-cnn",
+        "model_name", type=str, default="facebook/bart-large-cnn", help="like bart-large-cnn",
     )
     parser.add_argument(
         "--device", type=str, required=False, default=DEFAULT_DEVICE, help="cuda, cuda:1, cpu etc.",

examples/summarization/bart/test_bart_examples.py

@@ -129,7 +129,7 @@ class TestBartExamples(unittest.TestCase):
         summaries = ["A very interesting story about what I ate for lunch.", "Avocado, celery, turkey, coffee"]
         _dump_articles((tmp_dir / "train.source"), articles)
         _dump_articles((tmp_dir / "train.target"), summaries)
-        tokenizer = BartTokenizer.from_pretrained("bart-large")
+        tokenizer = BartTokenizer.from_pretrained("facebook/bart-large")
         max_len_source = max(len(tokenizer.encode(a)) for a in articles)
         max_len_target = max(len(tokenizer.encode(a)) for a in summaries)
         trunc_target = 4

src/transformers/configuration_ctrl.py

@@ -22,7 +22,7 @@ from .configuration_utils import PretrainedConfig
 
 logger = logging.getLogger(__name__)
 
-CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP = {"ctrl": "https://storage.googleapis.com/sf-ctrl/pytorch/ctrl-config.json"}
+CTRL_PRETRAINED_CONFIG_ARCHIVE_MAP = {"ctrl": "https://s3.amazonaws.com/models.huggingface.co/bert/ctrl-config.json"}
 
 
 class CTRLConfig(PretrainedConfig):

src/transformers/convert_bart_original_pytorch_checkpoint_to_pytorch.py

@@ -124,7 +124,7 @@ def convert_bart_checkpoint(checkpoint_path, pytorch_dump_folder_path, hf_checkp
         remove_ignore_keys_(state_dict)
         state_dict["shared.weight"] = state_dict["decoder.embed_tokens.weight"]
         fairseq_output = bart.extract_features(tokens)
-        if hf_checkpoint_name == "bart-large":
+        if hf_checkpoint_name == "facebook/bart-large":
             model = BartModel(config).eval()
             model.load_state_dict(state_dict)
             new_model_outputs = model(tokens).model[0]

src/transformers/pipelines.py

@@ -1633,7 +1633,7 @@ SUPPORTED_TASKS = {
         "impl": SummarizationPipeline,
         "tf": TFAutoModelWithLMHead if is_tf_available() else None,
         "pt": AutoModelWithLMHead if is_torch_available() else None,
-        "default": {"model": {"pt": "bart-large-cnn", "tf": "t5-small"}, "config": None, "tokenizer": None},
+        "default": {"model": {"pt": "facebook/bart-large-cnn", "tf": "t5-small"}, "config": None, "tokenizer": None},
     },
     "translation_en_to_fr": {
         "impl": TranslationPipeline,

src/transformers/tokenization_bart.py

@@ -25,7 +25,12 @@ logger = logging.getLogger(__name__)
 # vocab and merges same as roberta
 vocab_url = "https://s3.amazonaws.com/models.huggingface.co/bert/roberta-large-vocab.json"
 merges_url = "https://s3.amazonaws.com/models.huggingface.co/bert/roberta-large-merges.txt"
-_all_bart_models = ["bart-large", "bart-large-mnli", "bart-large-cnn", "bart-large-xsum"]
+_all_bart_models = [
+    "facebook/bart-large",
+    "facebook/bart-large-mnli",
+    "facebook/bart-large-cnn",
+    "facebook/bart-large-xsum",
+]
 
 
 class BartTokenizer(RobertaTokenizer):
@@ -37,7 +42,7 @@ class BartTokenizer(RobertaTokenizer):
     }
 
 
-_all_mbart_models = ["mbart-large-en-ro"]
+_all_mbart_models = ["facebook/mbart-large-en-ro"]
 SPM_URL = "https://s3.amazonaws.com/models.huggingface.co/bert/facebook/mbart-large-en-ro/sentence.bpe.model"
 
 

src/transformers/tokenization_flaubert.py

@@ -32,31 +32,31 @@ VOCAB_FILES_NAMES = {
 
 PRETRAINED_VOCAB_FILES_MAP = {
     "vocab_file": {
-        "flaubert-small-cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_small_cased/vocab.json",
-        "flaubert-base-uncased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_base_uncased/vocab.json",
-        "flaubert-base-cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_base_cased/vocab.json",
-        "flaubert-large-cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_large_cased/vocab.json",
+        "flaubert/flaubert_small_cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_small_cased/vocab.json",
+        "flaubert/flaubert_base_uncased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_base_uncased/vocab.json",
+        "flaubert/flaubert_base_cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_base_cased/vocab.json",
+        "flaubert/flaubert_large_cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_large_cased/vocab.json",
     },
     "merges_file": {
-        "flaubert-small-cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_small_cased/merges.txt",
-        "flaubert-base-uncased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_base_uncased/merges.txt",
-        "flaubert-base-cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_base_cased/merges.txt",
-        "flaubert-large-cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_large_cased/merges.txt",
+        "flaubert/flaubert_small_cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_small_cased/merges.txt",
+        "flaubert/flaubert_base_uncased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_base_uncased/merges.txt",
+        "flaubert/flaubert_base_cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_base_cased/merges.txt",
+        "flaubert/flaubert_large_cased": "https://s3.amazonaws.com/models.huggingface.co/bert/flaubert/flaubert_large_cased/merges.txt",
     },
 }
 
 PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
-    "flaubert-small-cased": 512,
-    "flaubert-base-uncased": 512,
-    "flaubert-base-cased": 512,
-    "flaubert-large-cased": 512,
+    "flaubert/flaubert_small_cased": 512,
+    "flaubert/flaubert_base_uncased": 512,
+    "flaubert/flaubert_base_cased": 512,
+    "flaubert/flaubert_large_cased": 512,
 }
 
 PRETRAINED_INIT_CONFIGURATION = {
-    "flaubert-small-cased": {"do_lowercase": False},
-    "flaubert-base-uncased": {"do_lowercase": True},
-    "flaubert-base-cased": {"do_lowercase": False},
-    "flaubert-large-cased": {"do_lowercase": False},
+    "flaubert/flaubert_small_cased": {"do_lowercase": False},
+    "flaubert/flaubert_base_uncased": {"do_lowercase": True},
+    "flaubert/flaubert_base_cased": {"do_lowercase": False},
+    "flaubert/flaubert_large_cased": {"do_lowercase": False},
 }
 
 

tests/test_modeling_bart.py

@@ -203,7 +203,7 @@ class BartTranslationTests(unittest.TestCase):
 
     @classmethod
     def setUpClass(cls):
-        checkpoint_name = "mbart-large-en-ro"
+        checkpoint_name = "facebook/mbart-large-en-ro"
         cls.tokenizer = MBartTokenizer.from_pretrained(checkpoint_name)
         cls.pad_token_id = 1
         net_input = {
@@ -230,7 +230,7 @@ class BartTranslationTests(unittest.TestCase):
     def model(self):
         """Only load the model if needed."""
         if self._model is None:
-            model = BartForConditionalGeneration.from_pretrained("mbart-large-en-ro")
+            model = BartForConditionalGeneration.from_pretrained("facebook/mbart-large-en-ro")
             self._model = model.to(torch_device)
         return self._model
 
@@ -427,7 +427,7 @@ class BartHeadTests(unittest.TestCase):
 
     @slow
     def test_tokenization(self):
-        tokenizer = BartTokenizer.from_pretrained("bart-large")
+        tokenizer = BartTokenizer.from_pretrained("facebook/bart-large")
         examples = [" Hello world", " DomDramg"]  # need leading spaces for equality
         fairseq_results = [
             torch.Tensor([0, 20920, 232, 2]),
@@ -519,7 +519,7 @@ TOLERANCE = 1e-4
 class BartModelIntegrationTests(unittest.TestCase):
     @slow
     def test_inference_no_head(self):
-        model = BartModel.from_pretrained("bart-large").to(torch_device)
+        model = BartModel.from_pretrained("facebook/bart-large").to(torch_device)
         input_ids = _long_tensor([[0, 31414, 232, 328, 740, 1140, 12695, 69, 46078, 1588, 2]])
         inputs_dict = prepare_bart_inputs_dict(model.config, input_ids)
         with torch.no_grad():
@@ -537,7 +537,7 @@ class BartModelIntegrationTests(unittest.TestCase):
         example_b = [0, 31414, 232, 328, 740, 1140, 69, 46078, 1588, 2, 1]
         input_ids = _long_tensor([[0, 31414, 232, 328, 740, 1140, 12695, 69, 46078, 1588, 2], example_b])
 
-        model = AutoModelForSequenceClassification.from_pretrained("bart-large-mnli").to(
+        model = AutoModelForSequenceClassification.from_pretrained("facebook/bart-large-mnli").to(
             torch_device
         )  # eval called in from_pre
         inputs_dict = prepare_bart_inputs_dict(model.config, input_ids)
@@ -567,9 +567,9 @@ class BartModelIntegrationTests(unittest.TestCase):
 
     @slow
     def test_xsum_summarization_same_as_fairseq(self):
-        model = BartForConditionalGeneration.from_pretrained("bart-large-xsum").to(torch_device)
+        model = BartForConditionalGeneration.from_pretrained("facebook/bart-large-xsum").to(torch_device)
         self.assertFalse(model.config.is_valid_mbart())
-        tok = BartTokenizer.from_pretrained("bart-large")
+        tok = BartTokenizer.from_pretrained("facebook/bart-large")
 
         PGE_ARTICLE = """ PG&E stated it scheduled the blackouts in response to forecasts for high winds amid dry conditions. The aim is to reduce the risk of wildfires. Nearly 800 thousand customers were scheduled to be affected by the shutoffs which were expected to last through at least midday tomorrow."""
         EXPECTED_SUMMARY = "California's largest power company has begun shutting off power to tens of thousands of homes and businesses in the state."
@@ -600,8 +600,8 @@ class BartModelIntegrationTests(unittest.TestCase):
 
     @slow
     def test_cnn_summarization_same_as_fairseq(self):
-        hf = BartForConditionalGeneration.from_pretrained("bart-large-cnn").to(torch_device)
-        tok = BartTokenizer.from_pretrained("bart-large")
+        hf = BartForConditionalGeneration.from_pretrained("facebook/bart-large-cnn").to(torch_device)
+        tok = BartTokenizer.from_pretrained("facebook/bart-large")
 
         FRANCE_ARTICLE = ' Marseille, France (CNN)The French prosecutor leading an investigation into the crash of Germanwings Flight 9525 insisted Wednesday that he was not aware of any video footage from on board the plane. Marseille prosecutor Brice Robin told CNN that "so far no videos were used in the crash investigation." He added, "A person who has such a video needs to immediately give it to the investigators." Robin\'s comments follow claims by two magazines, German daily Bild and French Paris Match, of a cell phone video showing the harrowing final seconds from on board Germanwings Flight 9525 as it crashed into the French Alps. All 150 on board were killed. Paris Match and Bild reported that the video was recovered from a phone at the wreckage site. The two publications described the supposed video, but did not post it on their websites. The publications said that they watched the video, which was found by a source close to the investigation. "One can hear cries of \'My God\' in several languages," Paris Match reported. "Metallic banging can also be heard more than three times, perhaps of the pilot trying to open the cockpit door with a heavy object.  Towards the end, after a heavy shake, stronger than the others, the screaming intensifies. Then nothing." "It is a very disturbing scene," said Julian Reichelt, editor-in-chief of Bild online. An official with France\'s accident investigation agency, the BEA, said the agency is not aware of any such video. Lt. Col. Jean-Marc Menichini, a French Gendarmerie spokesman in charge of communications on rescue efforts around the Germanwings crash site, told CNN that the reports were "completely wrong" and "unwarranted." Cell phones have been collected at the site, he said, but that they "hadn\'t been exploited yet." Menichini said he believed the cell phones would need to be sent to the Criminal Research Institute in Rosny sous-Bois, near Paris, in order to be analyzed by specialized technicians working hand-in-hand with investigators. But none of the cell phones found so far have been sent to the institute, Menichini said. Asked whether staff involved in the search could have leaked a memory card to the media, Menichini answered with a categorical "no." Reichelt told "Erin Burnett: Outfront" that he had watched the video and stood by the report, saying Bild and Paris Match are "very confident" that the clip is real. He noted that investigators only revealed they\'d recovered cell phones from the crash site after Bild and Paris Match published their reports. "That is something we did not know before. ... Overall we can say many things of the investigation weren\'t revealed by the investigation at the beginning," he said. What was mental state of Germanwings co-pilot? German airline Lufthansa confirmed Tuesday that co-pilot Andreas Lubitz had battled depression years before he took the controls of Germanwings Flight 9525, which he\'s accused of deliberately crashing last week in the French Alps. Lubitz told his Lufthansa flight training school in 2009 that he had a "previous episode of severe depression," the airline said Tuesday. Email correspondence between Lubitz and the school discovered in an internal investigation, Lufthansa said, included medical documents he submitted in connection with resuming his flight training. The announcement indicates that Lufthansa, the parent company of Germanwings, knew of Lubitz\'s battle with depression, allowed him to continue training and ultimately put him in the cockpit. Lufthansa, whose CEO Carsten Spohr previously said Lubitz was 100% fit to fly, described its statement Tuesday as a "swift and seamless clarification" and said it was sharing the information and documents -- including training and medical records -- with public prosecutors. Spohr traveled to the crash site Wednesday, where recovery teams have been working for the past week to recover human remains and plane debris scattered across a steep mountainside. He saw the crisis center set up in Seyne-les-Alpes, laid a wreath in the village of Le Vernet, closer to the crash site, where grieving families have left flowers at a simple stone memorial. Menichini told CNN late Tuesday that no visible human remains were left at the site but recovery teams would keep searching. French President Francois Hollande, speaking Tuesday, said that it should be possible to identify all the victims using DNA analysis by the end of the week, sooner than authorities had previously suggested. In the meantime, the recovery of the victims\' personal belongings will start Wednesday, Menichini said. Among those personal belongings could be more cell phones belonging to the 144 passengers and six crew on board. Check out the latest from our correspondents . The details about Lubitz\'s correspondence with the flight school during his training were among several developments as investigators continued to delve into what caused the crash and Lubitz\'s possible motive for downing the jet. A Lufthansa spokesperson told CNN on Tuesday that Lubitz had a valid medical certificate, had passed all his examinations and "held all the licenses required." Earlier, a spokesman for the prosecutor\'s office in Dusseldorf, Christoph Kumpa, said medical records reveal Lubitz suffered from suicidal tendencies at some point before his aviation career and underwent psychotherapy before he got his pilot\'s license. Kumpa emphasized there\'s no evidence suggesting Lubitz was suicidal or acting aggressively before the crash. Investigators are looking into whether Lubitz feared his medical condition would cause him to lose his pilot\'s license, a European government official briefed on the investigation told CNN on Tuesday. While flying was "a big part of his life," the source said, it\'s only one theory being considered. Another source, a law enforcement official briefed on the investigation, also told CNN that authorities believe the primary motive for Lubitz to bring down the plane was that he feared he would not be allowed to fly because of his medical problems. Lubitz\'s girlfriend told investigators he had seen an eye doctor and a neuropsychologist, both of whom deemed him unfit to work recently and concluded he had psychological issues, the European government official said. But no matter what details emerge about his previous mental health struggles, there\'s more to the story, said Brian Russell, a forensic psychologist. "Psychology can explain why somebody would turn rage inward on themselves about the fact that maybe they weren\'t going to keep doing their job and they\'re upset about that and so they\'re suicidal," he said. "But there is no mental illness that explains why somebody then feels entitled to also take that rage and turn it outward on 149 other people who had nothing to do with the person\'s problems." Germanwings crash compensation: What we know . Who was the captain of Germanwings Flight 9525? CNN\'s Margot Haddad reported from Marseille and Pamela Brown from Dusseldorf, while Laura Smith-Spark wrote from London. CNN\'s Frederik Pleitgen, Pamela Boykoff, Antonia Mortensen, Sandrine Amiel and Anna-Maja Rappard contributed to this report.'  # @noqa
         EXPECTED_SUMMARY_FRANCE = 'French prosecutor says he\'s not aware of any video footage from on board the plane. German daily Bild and French Paris Match claim to have found a cell phone video of the crash. A French Gendarmerie spokesman calls the reports "completely wrong" and "unwarranted" German airline Lufthansa confirms co-pilot Andreas Lubitz had battled depression.'