End of training

README.md

@@ -25,16 +25,16 @@ model-index:
     metrics:
     - name: Precision
       type: precision
-      value: 0.8641078838174274
+      value: 0.8750653423941454
     - name: Recall
       type: recall
-      value: 0.8904329235702833
+      value: 0.89470871191876
     - name: F1
       type: f1
-      value: 0.877072913924717
+      value: 0.8847780126849896
     - name: Accuracy
       type: accuracy
-      value: 0.9697581443217353
+      value: 0.9699164786446582
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -44,11 +44,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [FacebookAI/xlm-roberta-large](https://huggingface.co/FacebookAI/xlm-roberta-large) on the cnec dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1720
-- Precision: 0.8641
-- Recall: 0.8904
-- F1: 0.8771
-- Accuracy: 0.9698
+- Loss: 0.2020
+- Precision: 0.8751
+- Recall: 0.8947
+- F1: 0.8848
+- Accuracy: 0.9699
 
 ## Model description
 
@@ -79,16 +79,16 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.4839        | 1.0   | 581  | 0.1812          | 0.7650    | 0.8038 | 0.7839 | 0.9495   |
-| 0.2176        | 2.0   | 1162 | 0.1446          | 0.8225    | 0.8370 | 0.8297 | 0.9625   |
-| 0.1579        | 3.0   | 1743 | 0.1526          | 0.8164    | 0.8578 | 0.8366 | 0.9645   |
-| 0.1391        | 4.0   | 2324 | 0.1362          | 0.8495    | 0.8626 | 0.8560 | 0.9663   |
-| 0.1163        | 5.0   | 2905 | 0.1534          | 0.8456    | 0.8749 | 0.8600 | 0.9672   |
-| 0.0976        | 6.0   | 3486 | 0.1567          | 0.8454    | 0.8765 | 0.8607 | 0.9653   |
-| 0.0725        | 7.0   | 4067 | 0.1629          | 0.8580    | 0.8813 | 0.8695 | 0.9676   |
-| 0.0626        | 8.0   | 4648 | 0.1749          | 0.8664    | 0.8904 | 0.8782 | 0.9683   |
-| 0.0564        | 9.0   | 5229 | 0.1680          | 0.8649    | 0.8894 | 0.8769 | 0.9706   |
-| 0.049         | 10.0  | 5810 | 0.1720          | 0.8641    | 0.8904 | 0.8771 | 0.9698   |
+| 0.3776        | 1.0   | 581  | 0.1732          | 0.7868    | 0.8423 | 0.8136 | 0.9580   |
+| 0.1773        | 2.0   | 1162 | 0.1476          | 0.8243    | 0.8675 | 0.8453 | 0.9625   |
+| 0.127         | 3.0   | 1743 | 0.1522          | 0.8373    | 0.8691 | 0.8529 | 0.9654   |
+| 0.1057        | 4.0   | 2324 | 0.1516          | 0.8604    | 0.8728 | 0.8665 | 0.9665   |
+| 0.0852        | 5.0   | 2905 | 0.1555          | 0.8501    | 0.8883 | 0.8688 | 0.9700   |
+| 0.069         | 6.0   | 3486 | 0.1847          | 0.8637    | 0.8910 | 0.8771 | 0.9681   |
+| 0.0452        | 7.0   | 4067 | 0.1751          | 0.8666    | 0.8851 | 0.8757 | 0.9682   |
+| 0.0385        | 8.0   | 4648 | 0.1968          | 0.8626    | 0.8888 | 0.8755 | 0.9690   |
+| 0.0326        | 9.0   | 5229 | 0.1932          | 0.8717    | 0.8936 | 0.8826 | 0.9704   |
+| 0.026         | 10.0  | 5810 | 0.2020          | 0.8751    | 0.8947 | 0.8848 | 0.9699   |
 
 
 ### Framework versions

config.json

@@ -3,12 +3,12 @@
   "architectures": [
     "XLMRobertaForTokenClassification"
   ],
-  "attention_probs_dropout_prob": 0.25,
+  "attention_probs_dropout_prob": 0.2,
   "bos_token_id": 0,
   "classifier_dropout": null,
   "eos_token_id": 2,
   "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.25,
+  "hidden_dropout_prob": 0.2,
   "hidden_size": 1024,
   "id2label": {
     "0": "O",

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b73162d300cfa0dd216f5343b6d3bf565aec740f2b8bf03693beffc8161c530
+oid sha256:1b4f28b0626e84f9d3f45947f4f00d229f16b707199e727551cee10bd973aa1a
 size 2235473356

tmp-checkpoint-2910/config.json

@@ -0,0 +1,62 @@
+{
+  "_name_or_path": "FacebookAI/xlm-roberta-large",
+  "architectures": [
+    "XLMRobertaForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.15,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.3,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "O",
+    "1": "B-A",
+    "2": "I-A",
+    "3": "B-G",
+    "4": "I-G",
+    "5": "B-I",
+    "6": "I-I",
+    "7": "B-M",
+    "8": "I-M",
+    "9": "B-O",
+    "10": "I-O",
+    "11": "B-P",
+    "12": "I-P",
+    "13": "B-T",
+    "14": "I-T"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "B-A": 1,
+    "B-G": 3,
+    "B-I": 5,
+    "B-M": 7,
+    "B-O": 9,
+    "B-P": 11,
+    "B-T": 13,
+    "I-A": 2,
+    "I-G": 4,
+    "I-I": 6,
+    "I-M": 8,
+    "I-O": 10,
+    "I-P": 12,
+    "I-T": 14,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a6202687fe3aee68ea8ab8cea2ebaf5b94ee51f9a23262a8547aff52ab6b2dbc
+oid sha256:1c60a2cfe4ca61397b29fd56734d98f8d7ac8ab9e0eb13ffd59562de2b36795d
 size 4728