[Umberto] model shortcuts (#2661)

* [Umberto] model shortcuts

cc @loretoparisi @simonefrancia

see #2485

* Ensure that tokenizers will be correctly configured

src/transformers/configuration_camembert.py

@@ -25,6 +25,8 @@ logger = logging.getLogger(__name__)
 
 CAMEMBERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
     "camembert-base": "https://s3.amazonaws.com/models.huggingface.co/bert/camembert-base-config.json",
+    "umberto-commoncrawl-cased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/Musixmatch/umberto-commoncrawl-cased-v1/config.json",
+    "umberto-wikipedia-uncased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/Musixmatch/umberto-wikipedia-uncased-v1/config.json",
 }
 
 

src/transformers/modeling_camembert.py

@@ -33,6 +33,8 @@ logger = logging.getLogger(__name__)
 
 CAMEMBERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
     "camembert-base": "https://s3.amazonaws.com/models.huggingface.co/bert/camembert-base-pytorch_model.bin",
+    "umberto-commoncrawl-cased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/Musixmatch/umberto-commoncrawl-cased-v1/pytorch_model.bin",
+    "umberto-wikipedia-uncased-v1": "https://s3.amazonaws.com/models.huggingface.co/bert/Musixmatch/umberto-wikipedia-uncased-v1/pytorch_model.bin",
 }
 
 

src/transformers/tokenization_camembert.py

@@ -40,6 +40,13 @@ PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
     "camembert-base": None,
 }
 
+SHARED_MODEL_IDENTIFIERS = [
+    # Load with
+    # `tokenizer = AutoTokenizer.from_pretrained("username/pretrained_model")`
+    "Musixmatch/umberto-commoncrawl-cased-v1",
+    "Musixmatch/umberto-wikipedia-uncased-v1",
+]
+
 
 class CamembertTokenizer(PreTrainedTokenizer):
     """