DeFormer v2.0 de/dem/det/enda/ända support
Browse files- config.json +11 -4
- pytorch_model.bin +2 -2
- tokenizer_config.json +16 -1
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"architectures": [
|
| 4 |
"BertForTokenClassification"
|
| 5 |
],
|
|
@@ -12,17 +12,24 @@
|
|
| 12 |
"id2label": {
|
| 13 |
"0": "ord",
|
| 14 |
"1": "DE",
|
| 15 |
-
"2": "DEM"
|
|
|
|
|
|
|
|
|
|
| 16 |
},
|
| 17 |
"initializer_range": 0.02,
|
| 18 |
"intermediate_size": 3072,
|
| 19 |
"label2id": {
|
| 20 |
"DE": "1",
|
| 21 |
"DEM": "2",
|
| 22 |
-
"
|
|
|
|
|
|
|
|
|
|
| 23 |
},
|
| 24 |
"layer_norm_eps": 1e-12,
|
| 25 |
"max_position_embeddings": 512,
|
|
|
|
| 26 |
"model_type": "bert",
|
| 27 |
"num_attention_heads": 12,
|
| 28 |
"num_hidden_layers": 12,
|
|
@@ -34,4 +41,4 @@
|
|
| 34 |
"type_vocab_size": 2,
|
| 35 |
"use_cache": true,
|
| 36 |
"vocab_size": 50325
|
| 37 |
-
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "KBLab/bert-base-swedish-cased",
|
| 3 |
"architectures": [
|
| 4 |
"BertForTokenClassification"
|
| 5 |
],
|
|
|
|
| 12 |
"id2label": {
|
| 13 |
"0": "ord",
|
| 14 |
"1": "DE",
|
| 15 |
+
"2": "DEM",
|
| 16 |
+
"3": "DET",
|
| 17 |
+
"4": "ENDA",
|
| 18 |
+
"5": "\u00c4NDA"
|
| 19 |
},
|
| 20 |
"initializer_range": 0.02,
|
| 21 |
"intermediate_size": 3072,
|
| 22 |
"label2id": {
|
| 23 |
"DE": "1",
|
| 24 |
"DEM": "2",
|
| 25 |
+
"DET": "3",
|
| 26 |
+
"ENDA": "4",
|
| 27 |
+
"ord": "0",
|
| 28 |
+
"\u00c4NDA": "5"
|
| 29 |
},
|
| 30 |
"layer_norm_eps": 1e-12,
|
| 31 |
"max_position_embeddings": 512,
|
| 32 |
+
"model_max_length": 512,
|
| 33 |
"model_type": "bert",
|
| 34 |
"num_attention_heads": 12,
|
| 35 |
"num_hidden_layers": 12,
|
|
|
|
| 41 |
"type_vocab_size": 2,
|
| 42 |
"use_cache": true,
|
| 43 |
"vocab_size": 50325
|
| 44 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36c8f21e841ea95e7f491f1468a822931cf1d9399835542d1f136eb7463730bc
|
| 3 |
+
size 496503281
|
tokenizer_config.json
CHANGED
|
@@ -1 +1,16 @@
|
|
| 1 |
-
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_lower_case": false,
|
| 3 |
+
"unk_token": "[UNK]",
|
| 4 |
+
"sep_token": "[SEP]",
|
| 5 |
+
"pad_token": "[PAD]",
|
| 6 |
+
"cls_token": "[CLS]",
|
| 7 |
+
"mask_token": "[MASK]",
|
| 8 |
+
"tokenize_chinese_chars": true,
|
| 9 |
+
"strip_accents": false,
|
| 10 |
+
"special_tokens_map_file": "/home/faton/.cache/huggingface/transformers/04a84d173cafb30222a62efc0407878ebf2dc63e1871634cccbd4b68b923b031.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d",
|
| 11 |
+
"name_or_path": "KBLab/bert-base-swedish-cased",
|
| 12 |
+
"do_basic_tokenize": true,
|
| 13 |
+
"never_split": null,
|
| 14 |
+
"tokenizer_class": "BertTokenizer",
|
| 15 |
+
"model_max_length": 512
|
| 16 |
+
}
|