saattrupdan commited on
Commit
daef20d
1 Parent(s): 3a9917d
.gitignore ADDED
@@ -0,0 +1 @@
 
 
1
+ checkpoint-*/
README.md ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-4.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - null
7
+ model-index:
8
+ - name: nbailab-base-ner-scandi-unbalanced
9
+ results:
10
+ - task:
11
+ name: Token Classification
12
+ type: token-classification
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # nbailab-base-ner-scandi-unbalanced
19
+
20
+ This model is a fine-tuned version of [NbAiLab/nb-bert-base](https://huggingface.co/NbAiLab/nb-bert-base) on the None dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 0.0666
23
+ - Micro F1: 0.8693
24
+ - Micro F1 No Misc: 0.8925
25
+
26
+ ## Model description
27
+
28
+ More information needed
29
+
30
+ ## Intended uses & limitations
31
+
32
+ More information needed
33
+
34
+ ## Training and evaluation data
35
+
36
+ More information needed
37
+
38
+ ## Training procedure
39
+
40
+ ### Training hyperparameters
41
+
42
+ The following hyperparameters were used during training:
43
+ - learning_rate: 2e-05
44
+ - train_batch_size: 8
45
+ - eval_batch_size: 8
46
+ - seed: 42
47
+ - gradient_accumulation_steps: 4
48
+ - total_train_batch_size: 32
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: linear
51
+ - lr_scheduler_warmup_steps: 90135.90000000001
52
+ - num_epochs: 1000
53
+
54
+ ### Training results
55
+
56
+ | Training Loss | Epoch | Step | Validation Loss | Micro F1 | Micro F1 No Misc |
57
+ |:-------------:|:-----:|:-----:|:---------------:|:--------:|:----------------:|
58
+ | 0.6682 | 1.0 | 2816 | 0.0872 | 0.6916 | 0.7306 |
59
+ | 0.0684 | 2.0 | 5632 | 0.0464 | 0.8167 | 0.8538 |
60
+ | 0.0444 | 3.0 | 8448 | 0.0367 | 0.8485 | 0.8783 |
61
+ | 0.0349 | 4.0 | 11264 | 0.0316 | 0.8684 | 0.8920 |
62
+ | 0.0282 | 5.0 | 14080 | 0.0290 | 0.8820 | 0.9033 |
63
+ | 0.0231 | 6.0 | 16896 | 0.0283 | 0.8854 | 0.9060 |
64
+ | 0.0189 | 7.0 | 19712 | 0.0253 | 0.8964 | 0.9156 |
65
+ | 0.0155 | 8.0 | 22528 | 0.0260 | 0.9016 | 0.9201 |
66
+ | 0.0123 | 9.0 | 25344 | 0.0266 | 0.9059 | 0.9233 |
67
+ | 0.0098 | 10.0 | 28160 | 0.0280 | 0.9091 | 0.9279 |
68
+ | 0.008 | 11.0 | 30976 | 0.0309 | 0.9093 | 0.9287 |
69
+ | 0.0065 | 12.0 | 33792 | 0.0313 | 0.9103 | 0.9284 |
70
+ | 0.0053 | 13.0 | 36608 | 0.0322 | 0.9078 | 0.9257 |
71
+ | 0.0046 | 14.0 | 39424 | 0.0343 | 0.9075 | 0.9256 |
72
+
73
+
74
+ ### Framework versions
75
+
76
+ - Transformers 4.10.3
77
+ - Pytorch 1.9.0+cu102
78
+ - Datasets 1.12.1
79
+ - Tokenizers 0.10.3
config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "NbAiLab/nb-bert-base",
3
+ "architectures": [
4
+ "BertForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "directionality": "bidi",
9
+ "gradient_checkpointing": false,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "id2label": {
14
+ "0": "B-LOC",
15
+ "1": "I-LOC",
16
+ "2": "B-ORG",
17
+ "3": "I-ORG",
18
+ "4": "B-PER",
19
+ "5": "I-PER",
20
+ "6": "B-MISC",
21
+ "7": "I-MISC",
22
+ "8": "O"
23
+ },
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 3072,
26
+ "label2id": {
27
+ "B-LOC": 0,
28
+ "B-MISC": 6,
29
+ "B-ORG": 2,
30
+ "B-PER": 4,
31
+ "I-LOC": 1,
32
+ "I-MISC": 7,
33
+ "I-ORG": 3,
34
+ "I-PER": 5,
35
+ "O": 8
36
+ },
37
+ "layer_norm_eps": 1e-12,
38
+ "max_position_embeddings": 512,
39
+ "model_type": "bert",
40
+ "num_attention_heads": 12,
41
+ "num_hidden_layers": 12,
42
+ "pad_token_id": 0,
43
+ "pooler_fc_size": 768,
44
+ "pooler_num_attention_heads": 12,
45
+ "pooler_num_fc_layers": 3,
46
+ "pooler_size_per_head": 128,
47
+ "pooler_type": "first_token_transform",
48
+ "position_embedding_type": "absolute",
49
+ "torch_dtype": "float32",
50
+ "transformers_version": "4.10.3",
51
+ "type_vocab_size": 2,
52
+ "use_cache": true,
53
+ "vocab_size": 119547
54
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6431f1d8fa2383bef0a83ea660dec09fd15dbfb9766a6cc16347d5df42451bb
3
+ size 709162481
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"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, "special_tokens_map_file": null, "name_or_path": "NbAiLab/nb-bert-base", "do_basic_tokenize": true, "never_split": null, "tokenizer_class": "BertTokenizer"}
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2db68c7bf76cfd21c0f70c0ed31dbc2277bf061d0bf39ff2967bf2bc6d8d7dbe
3
+ size 2671
vocab.txt ADDED
The diff for this file is too large to render. See raw diff