Hugging Face's logo

afrodp95
/
distilbert-base-uncased-finetuned-job-skills-ner

Token Classification
Transformers
TensorFlow
distilbert
generated_from_keras_callback
Inference Endpoints
Model card Files Community
1
afrodp95 commited on
Commit
3dbd321
•
1 Parent(s): 7ac7844

Training in progress epoch 0

Browse files
Files changed (3) hide show
  1. README.md +10 -13
  2. config.json +24 -24
  3. tf_model.h5 +1 -1
README.md CHANGED
@@ -14,13 +14,13 @@ probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Train Loss: 0.1364
18
- - Validation Loss: 0.1546
19
- - Train Precision: 0.3254
20
- - Train Recall: 0.3330
21
- - Train F1: 0.3292
22
- - Train Accuracy: 0.9528
23
- - Epoch: 3
24
 
25
  ## Model description
26
 
@@ -39,17 +39,14 @@ More information needed
39
  ### Training hyperparameters
40
 
41
  The following hyperparameters were used during training:
42
- - optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 924, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
43
  - training_precision: float32
44
 
45
  ### Training results
46
 
47
  | Train Loss | Validation Loss | Train Precision | Train Recall | Train F1 | Train Accuracy | Epoch |
48
  |:----------:|:---------------:|:---------------:|:------------:|:--------:|:--------------:|:-----:|
49
- | 0.3414 | 0.2340 | 0.3032 | 0.1738 | 0.2210 | 0.9453 | 0 |
50
- | 0.1902 | 0.1801 | 0.2864 | 0.2734 | 0.2798 | 0.9504 | 1 |
51
- | 0.1525 | 0.1571 | 0.3365 | 0.2798 | 0.3055 | 0.9539 | 2 |
52
- | 0.1364 | 0.1546 | 0.3254 | 0.3330 | 0.3292 | 0.9528 | 3 |
53
 
54
 
55
  ### Framework versions
@@ -57,4 +54,4 @@ The following hyperparameters were used during training:
57
  - Transformers 4.24.0
58
  - TensorFlow 2.9.2
59
  - Datasets 2.6.1
60
- - Tokenizers 0.13.1
14
 
15
  This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Train Loss: 0.3257
18
+ - Validation Loss: 0.1935
19
+ - Train Precision: 0.3122
20
+ - Train Recall: 0.2144
21
+ - Train F1: 0.2542
22
+ - Train Accuracy: 0.9521
23
+ - Epoch: 0
24
 
25
  ## Model description
26
 
39
  ### Training hyperparameters
40
 
41
  The following hyperparameters were used during training:
42
+ - optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 3e-05, 'decay_steps': 1386, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
43
  - training_precision: float32
44
 
45
  ### Training results
46
 
47
  | Train Loss | Validation Loss | Train Precision | Train Recall | Train F1 | Train Accuracy | Epoch |
48
  |:----------:|:---------------:|:---------------:|:------------:|:--------:|:--------------:|:-----:|
49
+ | 0.3257 | 0.1935 | 0.3122 | 0.2144 | 0.2542 | 0.9521 | 0 |
 
 
 
50
 
51
 
52
  ### Framework versions
54
  - Transformers 4.24.0
55
  - TensorFlow 2.9.2
56
  - Datasets 2.6.1
57
+ - Tokenizers 0.13.2
config.json CHANGED
@@ -10,34 +10,34 @@
10
  "hidden_dim": 3072,
11
  "id2label": {
12
  "0": "O",
13
- "1": "B-competencia-tecnica",
14
- "2": "I-competencia-tecnica",
15
- "3": "L-competencia-tecnica",
16
- "4": "U-competencia-tecnica",
17
- "5": "B-competencia-socioemocional",
18
- "6": "I-competencia-socioemocional",
19
- "7": "L-competencia-socioemocional",
20
- "8": "U-competencia-socioemocional",
21
- "9": "B-competencia-digital",
22
- "10": "I-competencia-digital",
23
- "11": "L-competencia-digital",
24
- "12": "U-competencia-digital"
25
  },
26
  "initializer_range": 0.02,
27
  "label2id": {
28
- "B-competencia-digital": 9,
29
- "B-competencia-socioemocional": 5,
30
- "B-competencia-tecnica": 1,
31
- "I-competencia-digital": 10,
32
- "I-competencia-socioemocional": 6,
33
- "I-competencia-tecnica": 2,
34
- "L-competencia-digital": 11,
35
- "L-competencia-socioemocional": 7,
36
- "L-competencia-tecnica": 3,
37
  "O": 0,
38
- "U-competencia-digital": 12,
39
- "U-competencia-socioemocional": 8,
40
- "U-competencia-tecnica": 4
41
  },
42
  "max_position_embeddings": 512,
43
  "model_type": "distilbert",
10
  "hidden_dim": 3072,
11
  "id2label": {
12
  "0": "O",
13
+ "1": "B-competencia-socioemocional",
14
+ "2": "I-competencia-socioemocional",
15
+ "3": "L-competencia-socioemocional",
16
+ "4": "U-competencia-socioemocional",
17
+ "5": "B-competencia-digital",
18
+ "6": "I-competencia-digital",
19
+ "7": "L-competencia-digital",
20
+ "8": "U-competencia-digital",
21
+ "9": "B-competencia-tecnica",
22
+ "10": "I-competencia-tecnica",
23
+ "11": "L-competencia-tecnica",
24
+ "12": "U-competencia-tecnica"
25
  },
26
  "initializer_range": 0.02,
27
  "label2id": {
28
+ "B-competencia-digital": 5,
29
+ "B-competencia-socioemocional": 1,
30
+ "B-competencia-tecnica": 9,
31
+ "I-competencia-digital": 6,
32
+ "I-competencia-socioemocional": 2,
33
+ "I-competencia-tecnica": 10,
34
+ "L-competencia-digital": 7,
35
+ "L-competencia-socioemocional": 3,
36
+ "L-competencia-tecnica": 11,
37
  "O": 0,
38
+ "U-competencia-digital": 8,
39
+ "U-competencia-socioemocional": 4,
40
+ "U-competencia-tecnica": 12
41
  },
42
  "max_position_embeddings": 512,
43
  "model_type": "distilbert",
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d2d1a710e0484f9d4bc3c90aa816637d18c86dde5c4a267c9a7b1f9910740be3
3
  size 265618792
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:959d35ee974bd20752ec9e0df93c4d18d69f991b9a322bc60a633a9ce44ad8de
3
  size 265618792