Hugging Face's logo

abishekcodes
/
bert-new-ner

Token Classification
Transformers
TensorBoard
Safetensors
bert
Generated from Trainer
Model card Files
xet
Metrics Community
abishekcodes commited on
Commit
b0f54b2
·
verified ·
1 Parent(s): 5fe4558

Model save

Browse files
Files changed (5) hide show
  1. README.md +11 -12
  2. best/config.json +26 -32
  3. best/model.safetensors +2 -2
  4. best/training_args.bin +1 -1
  5. model.safetensors +1 -1
README.md CHANGED
@@ -8,7 +8,6 @@ metrics:
8
  - precision
9
  - recall
10
  - f1
11
- - accuracy
12
  model-index:
13
  - name: bert-new-ner
14
  results: []
@@ -19,13 +18,12 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  # bert-new-ner
21
 
22
- This model is a fine-tuned version of [google-bert/bert-base-uncased](https://huggingface.co/google-bert/bert-base-uncased) on an unknown dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.0476
25
- - Precision: 0.9310
26
- - Recall: 0.9472
27
- - F1: 0.9390
28
- - Accuracy: 0.9390
29
 
30
  ## Model description
31
 
@@ -51,14 +49,15 @@ The following hyperparameters were used during training:
51
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
52
  - lr_scheduler_type: linear
53
  - num_epochs: 3
 
54
 
55
  ### Training results
56
 
57
- | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
58
- |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
59
- | 0.0382 | 1.0 | 1064 | 0.0537 | 0.9162 | 0.9273 | 0.9217 | 0.9217 |
60
- | 0.018 | 2.0 | 2128 | 0.0466 | 0.9253 | 0.9480 | 0.9365 | 0.9365 |
61
- | 0.0109 | 3.0 | 3192 | 0.0476 | 0.9310 | 0.9472 | 0.9390 | 0.9390 |
62
 
63
 
64
  ### Framework versions
 
8
  - precision
9
  - recall
10
  - f1
 
11
  model-index:
12
  - name: bert-new-ner
13
  results: []
 
18
 
19
  # bert-new-ner
20
 
21
+ This model is a fine-tuned version of [google-bert/bert-base-uncased](https://huggingface.co/google-bert/bert-base-uncased) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.0246
24
+ - Precision: 0.9645
25
+ - Recall: 0.9682
26
+ - F1: 0.9664
 
27
 
28
  ## Model description
29
 
 
49
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
50
  - lr_scheduler_type: linear
51
  - num_epochs: 3
52
+ - mixed_precision_training: Native AMP
53
 
54
  ### Training results
55
 
56
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 |
57
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|
58
+ | 0.0227 | 1.0 | 1002 | 0.0263 | 0.9540 | 0.9614 | 0.9577 |
59
+ | 0.0125 | 2.0 | 2004 | 0.0237 | 0.9554 | 0.9720 | 0.9637 |
60
+ | 0.0064 | 3.0 | 3006 | 0.0246 | 0.9645 | 0.9682 | 0.9664 |
61
 
62
 
63
  ### Framework versions
best/config.json CHANGED
@@ -9,46 +9,40 @@
9
  "hidden_dropout_prob": 0.1,
10
  "hidden_size": 768,
11
  "id2label": {
12
- "0": "B-AADHAAR",
13
- "1": "B-LOC",
14
- "2": "B-MISC",
15
  "3": "B-ORG",
16
- "4": "B-PAN",
17
- "5": "B-PER",
18
- "6": "B-PERSON",
19
  "7": "B-PHONE_NUMBER",
20
- "8": "B-STREET_ADDRESS",
21
- "9": "I-AADHAAR",
22
- "10": "I-LOC",
23
- "11": "I-MISC",
24
- "12": "I-ORG",
25
- "13": "I-PER",
26
- "14": "I-PERSON",
27
- "15": "I-PHONE_NUMBER",
28
- "16": "I-STREET_ADDRESS",
29
- "17": "O"
30
  },
31
  "initializer_range": 0.02,
32
  "intermediate_size": 3072,
33
  "label2id": {
34
- "B-AADHAAR": 0,
35
- "B-LOC": 1,
36
- "B-MISC": 2,
37
  "B-ORG": 3,
38
- "B-PAN": 4,
39
- "B-PER": 5,
40
- "B-PERSON": 6,
41
  "B-PHONE_NUMBER": 7,
42
- "B-STREET_ADDRESS": 8,
43
- "I-AADHAAR": 9,
44
- "I-LOC": 10,
45
- "I-MISC": 11,
46
- "I-ORG": 12,
47
- "I-PER": 13,
48
- "I-PERSON": 14,
49
- "I-PHONE_NUMBER": 15,
50
- "I-STREET_ADDRESS": 16,
51
- "O": 17
52
  },
53
  "layer_norm_eps": 1e-12,
54
  "max_position_embeddings": 512,
 
9
  "hidden_dropout_prob": 0.1,
10
  "hidden_size": 768,
11
  "id2label": {
12
+ "0": "O",
13
+ "1": "B-PERSON",
14
+ "2": "I-PERSON",
15
  "3": "B-ORG",
16
+ "4": "I-ORG",
17
+ "5": "B-LOC",
18
+ "6": "I-LOC",
19
  "7": "B-PHONE_NUMBER",
20
+ "8": "I-PHONE_NUMBER",
21
+ "9": "B-STREET_ADDRESS",
22
+ "10": "I-STREET_ADDRESS",
23
+ "11": "B-AADHAAR",
24
+ "12": "I-AADHAAR",
25
+ "13": "B-PAN",
26
+ "14": "I-PAN"
 
 
 
27
  },
28
  "initializer_range": 0.02,
29
  "intermediate_size": 3072,
30
  "label2id": {
31
+ "B-AADHAAR": 11,
32
+ "B-LOC": 5,
 
33
  "B-ORG": 3,
34
+ "B-PAN": 13,
35
+ "B-PERSON": 1,
 
36
  "B-PHONE_NUMBER": 7,
37
+ "B-STREET_ADDRESS": 9,
38
+ "I-AADHAAR": 12,
39
+ "I-LOC": 6,
40
+ "I-ORG": 4,
41
+ "I-PAN": 14,
42
+ "I-PERSON": 2,
43
+ "I-PHONE_NUMBER": 8,
44
+ "I-STREET_ADDRESS": 10,
45
+ "O": 0
 
46
  },
47
  "layer_norm_eps": 1e-12,
48
  "max_position_embeddings": 512,
best/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:95fccfa4aafeb3332324de28f0208e9576bb2d7a6c8f10763dd87ea273265e36
3
- size 435645304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8965bd3215149ef397ca5dddaf8598191cc7bf431cd4a62edec9689ae14342e
3
+ size 435636076
best/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:004bf963fa13d483f1a70ddf498ae6decd0a745941871b5ab66d6c226b472797
3
  size 5841
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5dcbce359a20f1852c98e57ea984140420a75c3af4ac918c0210a7e6751e50d
3
  size 5841
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cf7eead8eaa2160c8073d3dc8ad2972bd5f1422e6e66da3d7c17cd94ff7fbd37
3
  size 435636076
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8965bd3215149ef397ca5dddaf8598191cc7bf431cd4a62edec9689ae14342e
3
  size 435636076