JonatanGk commited on
Commit
ac71aac
1 Parent(s): fb96c2c

Update labels

Browse files
Files changed (2) hide show
  1. README.md +9 -17
  2. config.json +8 -0
README.md CHANGED
@@ -1,13 +1,11 @@
1
  ---
2
- license: apache-2.0
3
  tags:
4
  - generated_from_trainer
5
- datasets:
6
- - null
7
  metrics:
8
  - accuracy
9
  model-index:
10
- - name: roberta-base-bne-finetuned-mnli
11
  results:
12
  - task:
13
  name: Text Classification
@@ -18,27 +16,18 @@ model-index:
18
  value: 0.9607097303206997
19
  ---
20
 
21
- <!-- This model card has been generated automatically according to the information the Trainer had access to. You
22
- should probably proofread and complete it, then remove this comment. -->
23
 
24
- # roberta-base-bne-finetuned-mnli
25
 
26
- This model is a fine-tuned version of [BSC-TeMU/roberta-base-bne](https://huggingface.co/BSC-TeMU/roberta-base-bne) on the None dataset.
27
  It achieves the following results on the evaluation set:
 
28
  - Loss: 0.1657
29
  - Accuracy: 0.9607
30
 
31
- ## Model description
32
-
33
- More information needed
34
-
35
- ## Intended uses & limitations
36
-
37
- More information needed
38
-
39
  ## Training and evaluation data
40
 
41
- More information needed
42
 
43
  ## Training procedure
44
 
@@ -55,6 +44,8 @@ The following hyperparameters were used during training:
55
 
56
  ### Training results
57
 
 
 
58
  | Training Loss | Epoch | Step | Accuracy | Validation Loss |
59
  |:-------------:|:-----:|:-----:|:--------:|:---------------:|
60
  | 0.1512 | 1.0 | 22227 | 0.9501 | 0.1418 |
@@ -62,6 +53,7 @@ The following hyperparameters were used during training:
62
  | 0.0973 | 3.0 | 66681 | 0.9594 | 0.1397 |
63
  | 0.0658 | 4.0 | 88908 | 0.9607 | 0.1657 |
64
 
 
65
 
66
  ### Framework versions
67
 
 
1
  ---
2
+ language: es
3
  tags:
4
  - generated_from_trainer
 
 
5
  metrics:
6
  - accuracy
7
  model-index:
8
+ - name: roberta-base-bne-finetuned-ciberbullying-spanish
9
  results:
10
  - task:
11
  name: Text Classification
 
16
  value: 0.9607097303206997
17
  ---
18
 
19
+ # roberta-base-bne-finetuned-ciberbullying-spanish
 
20
 
21
+ This model is a fine-tuned version of [BSC-TeMU/roberta-base-bne](https://huggingface.co/BSC-TeMU/roberta-base-bne) on the dataset generated scrapping all social networks (Twitter, Youtube ...) to detect ciberbullying on Spanish.
22
 
 
23
  It achieves the following results on the evaluation set:
24
+
25
  - Loss: 0.1657
26
  - Accuracy: 0.9607
27
 
 
 
 
 
 
 
 
 
28
  ## Training and evaluation data
29
 
30
+ We use the concatenation from multiple datasets generated scrapping social networks (Twitter,Youtube,Discord...) to fine-tune this model. The total number of sentence pairs is above 360k sentences.
31
 
32
  ## Training procedure
33
 
 
44
 
45
  ### Training results
46
 
47
+ <details>
48
+
49
  | Training Loss | Epoch | Step | Accuracy | Validation Loss |
50
  |:-------------:|:-----:|:-----:|:--------:|:---------------:|
51
  | 0.1512 | 1.0 | 22227 | 0.9501 | 0.1418 |
 
53
  | 0.0973 | 3.0 | 66681 | 0.9594 | 0.1397 |
54
  | 0.0658 | 4.0 | 88908 | 0.9607 | 0.1657 |
55
 
56
+ </details>
57
 
58
  ### Framework versions
59
 
config.json CHANGED
@@ -3,6 +3,14 @@
3
  "architectures": [
4
  "RobertaForSequenceClassification"
5
  ],
 
 
 
 
 
 
 
 
6
  "attention_probs_dropout_prob": 0.0,
7
  "bos_token_id": 0,
8
  "classifier_dropout": null,
 
3
  "architectures": [
4
  "RobertaForSequenceClassification"
5
  ],
6
+ "id2label": {
7
+ "0": "Not_bullying",
8
+ "1": "Bullying"
9
+ },
10
+ "label2id": {
11
+ "Not_bullying": 0,
12
+ "Bullying": 1
13
+ },
14
  "attention_probs_dropout_prob": 0.0,
15
  "bos_token_id": 0,
16
  "classifier_dropout": null,