Commit 1

README.md

@@ -1,8 +1,10 @@
 ---
-license: apache-2.0
-base_model: bert-base-cased
+license: mit
+base_model: FacebookAI/roberta-large
 tags:
 - generated_from_trainer
+datasets:
+- few-nerd
 metrics:
 - precision
 - recall
@@ -10,7 +12,29 @@ metrics:
 - accuracy
 model-index:
 - name: bert-finetuned-ner
-  results: []
+  results:
+  - task:
+      name: Token Classification
+      type: token-classification
+    dataset:
+      name: few-nerd
+      type: few-nerd
+      config: supervised
+      split: validation
+      args: supervised
+    metrics:
+    - name: Precision
+      type: precision
+      value: 0.7833402370948971
+    - name: Recall
+      type: recall
+      value: 0.8147760612215589
+    - name: F1
+      type: f1
+      value: 0.798748969206943
+    - name: Accuracy
+      type: accuracy
+      value: 0.9425415670481714
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -18,13 +42,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert-finetuned-ner
 
-This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on an unknown dataset.
+This model is a fine-tuned version of [FacebookAI/roberta-large](https://huggingface.co/FacebookAI/roberta-large) on the few-nerd dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2040
-- Precision: 0.8009
-- Recall: 0.8341
-- F1: 0.8171
-- Accuracy: 0.9414
+- Loss: 0.2154
+- Precision: 0.7833
+- Recall: 0.8148
+- F1: 0.7987
+- Accuracy: 0.9425
 
 ## Model description
 
@@ -44,8 +68,8 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 8
-- eval_batch_size: 8
+- train_batch_size: 4
+- eval_batch_size: 4
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
@@ -55,14 +79,14 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step  | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.1983        | 1.0   | 16471 | 0.1868          | 0.7919    | 0.8142 | 0.8029 | 0.9382   |
-| 0.1527        | 2.0   | 32942 | 0.1889          | 0.7951    | 0.8344 | 0.8143 | 0.9406   |
-| 0.1142        | 3.0   | 49413 | 0.2040          | 0.8009    | 0.8341 | 0.8171 | 0.9414   |
+| 0.1957        | 1.0   | 32942 | 0.1963          | 0.7533    | 0.7992 | 0.7756 | 0.9386   |
+| 0.1596        | 2.0   | 65884 | 0.2025          | 0.7768    | 0.8063 | 0.7913 | 0.9416   |
+| 0.1207        | 3.0   | 98826 | 0.2154          | 0.7833    | 0.8148 | 0.7987 | 0.9425   |
 
 
 ### Framework versions
 
-- Transformers 4.37.0
-- Pytorch 2.1.2+cu118
-- Datasets 2.16.1
-- Tokenizers 0.15.1
+- Transformers 4.35.2
+- Pytorch 2.1.1+cu118
+- Datasets 2.15.0
+- Tokenizers 0.15.0

config.json

@@ -1,48 +1,49 @@
 {
-  "_name_or_path": "bert-base-cased",
+  "_name_or_path": "FacebookAI/roberta-large",
   "architectures": [
-    "BertForTokenClassification"
+    "RobertaForTokenClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
   "classifier_dropout": null,
-  "gradient_checkpointing": false,
+  "eos_token_id": 2,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "hidden_size": 1024,
   "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"
+    "1": "art",
+    "2": "building",
+    "3": "event",
+    "4": "location",
+    "5": "organization",
+    "6": "other",
+    "7": "person",
+    "8": "product"
   },
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "label2id": {
-    "B-LOC": 5,
-    "B-MISC": 7,
-    "B-ORG": 3,
-    "B-PER": 1,
-    "I-LOC": 6,
-    "I-MISC": 8,
-    "I-ORG": 4,
-    "I-PER": 2,
-    "O": 0
+    "O": "0",
+    "art": "1",
+    "building": "2",
+    "event": "3",
+    "location": "4",
+    "organization": "5",
+    "other": "6",
+    "person": "7",
+    "product": "8"
   },
-  "layer_norm_eps": 1e-12,
-  "max_position_embeddings": 512,
-  "model_type": "bert",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
-  "pad_token_id": 0,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 1,
   "position_embedding_type": "absolute",
   "torch_dtype": "float32",
   "transformers_version": "4.35.2",
-  "type_vocab_size": 2,
+  "type_vocab_size": 1,
   "use_cache": true,
-  "vocab_size": 28996
+  "vocab_size": 50265
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19f938b42032d26b8f4e1b260fe19da611b448e9e00260eee784da5cbc974e0b
-size 430929740
+oid sha256:7e57fc58e48e854e51bbd243cc40534a8dad25924f828725bad0a8d1181704c0
+size 1417325484

special_tokens_map.json

@@ -1,7 +1,15 @@
 {
-  "cls_token": "[CLS]",
-  "mask_token": "[MASK]",
-  "pad_token": "[PAD]",
-  "sep_token": "[SEP]",
-  "unk_token": "[UNK]"
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
 }

tokenizer_config.json

@@ -1,55 +1,57 @@
 {
+  "add_prefix_space": true,
   "added_tokens_decoder": {
     "0": {
-      "content": "[PAD]",
+      "content": "<s>",
       "lstrip": false,
-      "normalized": false,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
-    "100": {
-      "content": "[UNK]",
+    "1": {
+      "content": "<pad>",
       "lstrip": false,
-      "normalized": false,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
-    "101": {
-      "content": "[CLS]",
+    "2": {
+      "content": "</s>",
       "lstrip": false,
-      "normalized": false,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
-    "102": {
-      "content": "[SEP]",
+    "3": {
+      "content": "<unk>",
       "lstrip": false,
-      "normalized": false,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
     },
-    "103": {
-      "content": "[MASK]",
-      "lstrip": false,
+    "50264": {
+      "content": "<mask>",
+      "lstrip": true,
       "normalized": false,
       "rstrip": false,
       "single_word": false,
       "special": true
     }
   },
+  "bos_token": "<s>",
   "clean_up_tokenization_spaces": true,
-  "cls_token": "[CLS]",
-  "do_lower_case": false,
-  "mask_token": "[MASK]",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "mask_token": "<mask>",
   "model_max_length": 512,
-  "pad_token": "[PAD]",
-  "sep_token": "[SEP]",
-  "strip_accents": null,
-  "tokenize_chinese_chars": true,
-  "tokenizer_class": "BertTokenizer",
-  "unk_token": "[UNK]"
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:509f66ecb80beab84acc8f97d675c7e5f93b58aa9a7f93546401917b193639b8
+oid sha256:8a45f7688d7fa188cd32f8cc229a17aec1b0e243797ee4689e48ac4890ba78e3
 size 4600