updated model from notebook

README.md

@@ -16,11 +16,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model was trained from scratch on an unkown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0484
-- Precision: 0.9340
-- Recall: 0.9413
-- F1: 0.9376
-- Accuracy: 0.9875
+- Loss: 0.0388
+- Precision: 0.9360
+- Recall: 0.9458
+- F1: 0.9409
+- Accuracy: 0.9902
 
 ## Model description
 
@@ -51,9 +51,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.1931        | 1.0   | 878  | 0.0518          | 0.9146    | 0.9276 | 0.9210 | 0.9852   |
-| 0.0389        | 2.0   | 1756 | 0.0470          | 0.9261    | 0.9389 | 0.9325 | 0.9870   |
-| 0.0228        | 3.0   | 2634 | 0.0484          | 0.9340    | 0.9413 | 0.9376 | 0.9875   |
+| 0.1653        | 1.0   | 878  | 0.0465          | 0.9267    | 0.9300 | 0.9283 | 0.9883   |
+| 0.0322        | 2.0   | 1756 | 0.0404          | 0.9360    | 0.9431 | 0.9396 | 0.9897   |
+| 0.0185        | 3.0   | 2634 | 0.0388          | 0.9360    | 0.9458 | 0.9409 | 0.9902   |
 
 
 ### Framework versions

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "distilbert-base-uncased",
+  "_name_or_path": "distilbert-base-multilingual-cased",
   "activation": "gelu",
   "architectures": [
     "DistilBertForTokenClassification"
@@ -9,15 +9,15 @@
   "dropout": 0.1,
   "hidden_dim": 3072,
   "id2label": {
-    "0": "O",
-    "1": "B-PER",
-    "2": "I-PER",
-    "3": "B-ORG",
-    "4": "I-ORG",
-    "5": "B-LOC",
-    "6": "I-LOC",
-    "7": "B-MISC",
-    "8": "I-MISC"
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4",
+    "5": "LABEL_5",
+    "6": "LABEL_6",
+    "7": "LABEL_7",
+    "8": "LABEL_8"
   },
   "initializer_range": 0.02,
   "label2id": {
@@ -35,11 +35,12 @@
   "model_type": "distilbert",
   "n_heads": 12,
   "n_layers": 6,
+  "output_past": true,
   "pad_token_id": 0,
   "qa_dropout": 0.1,
   "seq_classif_dropout": 0.2,
   "sinusoidal_pos_embds": false,
   "tie_weights_": true,
   "transformers_version": "4.6.1",
-  "vocab_size": 30522
+  "vocab_size": 119547
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76b8f43b516f5f0340d896894e4aa0f3f903752bac218e5c8d1b8477a853a7c3
-size 265521605
+oid sha256:36de80edacc15d43b97c4284f99a5c39d304e498676d7cfcbb52cf0a7f48a045
+size 539006405

tokenizer_config.json

@@ -1 +1 @@
-{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "distilbert-base-uncased"}
+{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "distilbert-base-multilingual-cased"}