Lagadro commited on
Commit
6da5ee1
1 Parent(s): 2821a7f

Upload TFBertForTokenClassification

Browse files
Files changed (3) hide show
  1. README.md +16 -67
  2. config.json +42 -0
  3. tf_model.h5 +3 -0
README.md CHANGED
@@ -1,71 +1,21 @@
1
  ---
2
- language:
3
- - tr
4
  license: mit
5
  base_model: dbmdz/bert-base-turkish-cased
6
- datasets:
7
- - wikiann
8
- - tr
9
- metrics:
10
- - precision
11
- - recall
12
- - f1
13
- - accuracy
14
  model-index:
15
- - name: bert-base-turkish-cased-None
16
- results:
17
- - task:
18
- name: Token Classification
19
- type: token-classification
20
- dataset:
21
- name: wikiann
22
- type: wikiann
23
- args: default
24
- metrics:
25
- - name: precision
26
- type: precision
27
- value: 0.9026122547249308
28
- - name: recall
29
- type: recall
30
- value: 0.9218096877305139
31
- - name: f1
32
- type: f1
33
- value: 0.912109968979989
34
- - name: accuracy
35
- type: accuracy
36
- value: 0.9604539478979423
37
- - task:
38
- name: Token Classification
39
- type: token-classification
40
- dataset:
41
- name: tr
42
- type: tr
43
- metrics:
44
- - name: precision
45
- type: precision
46
- value: 0.9026122547249308
47
- - name: recall
48
- type: recall
49
- value: 0.9218096877305139
50
- - name: f1
51
- type: f1
52
- value: 0.912109968979989
53
- - name: accuracy
54
- type: accuracy
55
- value: 0.9604539478979423
56
  ---
57
 
58
- <!-- This model card has been generated automatically according to the information the Trainer had access to. You
59
- should probably proofread and complete it, then remove this comment. -->
60
 
61
- # bert-base-turkish-cased-None
62
 
63
- This model is a fine-tuned version of [dbmdz/bert-base-turkish-cased](https://huggingface.co/dbmdz/bert-base-turkish-cased) on the wikiann and the tr datasets.
64
  It achieves the following results on the evaluation set:
65
- - precision: 0.9026
66
- - recall: 0.9218
67
- - f1: 0.9121
68
- - accuracy: 0.9605
69
 
70
  ## Model description
71
 
@@ -84,17 +34,16 @@ More information needed
84
  ### Training hyperparameters
85
 
86
  The following hyperparameters were used during training:
87
- - num_train_epochs: 5
88
- - train_batch_size: 16
89
- - eval_batch_size: 32
90
- - learning_rate: 2e-05
91
- - weight_decay_rate: 0.01
92
- - num_warmup_steps: 0
93
- - fp16: True
94
 
95
  ### Framework versions
96
 
97
  - Transformers 4.38.2
98
- - Pytorch 2.1.0+cu121
99
  - Datasets 2.18.0
100
  - Tokenizers 0.15.2
 
1
  ---
 
 
2
  license: mit
3
  base_model: dbmdz/bert-base-turkish-cased
4
+ tags:
5
+ - generated_from_keras_callback
 
 
 
 
 
 
6
  model-index:
7
+ - name: bert-base-turkish-cased-ner
8
+ results: []
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9
  ---
10
 
11
+ <!-- This model card has been generated automatically according to the information Keras had access to. You should
12
+ probably proofread and complete it, then remove this comment. -->
13
 
14
+ # bert-base-turkish-cased-ner
15
 
16
+ This model is a fine-tuned version of [dbmdz/bert-base-turkish-cased](https://huggingface.co/dbmdz/bert-base-turkish-cased) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+
 
 
 
19
 
20
  ## Model description
21
 
 
34
  ### Training hyperparameters
35
 
36
  The following hyperparameters were used during training:
37
+ - optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 5315, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
38
+ - training_precision: float32
39
+
40
+ ### Training results
41
+
42
+
 
43
 
44
  ### Framework versions
45
 
46
  - Transformers 4.38.2
47
+ - TensorFlow 2.15.0
48
  - Datasets 2.18.0
49
  - Tokenizers 0.15.2
config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "dbmdz/bert-base-turkish-cased",
3
+ "architectures": [
4
+ "BertForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "O",
13
+ "1": "B-PER",
14
+ "2": "I-PER",
15
+ "3": "B-ORG",
16
+ "4": "I-ORG",
17
+ "5": "B-LOC",
18
+ "6": "I-LOC"
19
+ },
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 3072,
22
+ "label2id": {
23
+ "B-LOC": "5",
24
+ "B-ORG": "3",
25
+ "B-PER": "1",
26
+ "I-LOC": "6",
27
+ "I-ORG": "4",
28
+ "I-PER": "2",
29
+ "O": "0"
30
+ },
31
+ "layer_norm_eps": 1e-12,
32
+ "max_position_embeddings": 512,
33
+ "model_type": "bert",
34
+ "num_attention_heads": 12,
35
+ "num_hidden_layers": 12,
36
+ "pad_token_id": 0,
37
+ "position_embedding_type": "absolute",
38
+ "transformers_version": "4.38.2",
39
+ "type_vocab_size": 2,
40
+ "use_cache": true,
41
+ "vocab_size": 32000
42
+ }
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6190a06d3ad1874aef34cf8fb9383c2b8e1aa059351b7e4eb785d92b28ff5ed6
3
+ size 440401892