Hugging Face's logo

JFrediani
/
mBERT-base-offensive

Text Classification
Transformers
TensorBoard
Safetensors
bert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
JFrediani commited on
Commit
79ca2eb
·
verified ·
1 Parent(s): d16a386

BERT-L-offensive

Browse files
Files changed (7) hide show
  1. README.md +20 -20
  2. config.json +5 -5
  3. model.safetensors +2 -2
  4. runs/Jul20_19-12-56_85df0c1db32e/events.out.tfevents.1721515789.85df0c1db32e.2808.1 +2 -2
  5. runs/Jul20_23-10-58_85df0c1db32e/events.out.tfevents.1721517059.85df0c1db32e.61311.0 +3 -0
  6. runs/Jul20_23-10-58_85df0c1db32e/events.out.tfevents.1721521186.85df0c1db32e.61311.1 +3 -0
  7. training_args.bin +1 -1
README.md CHANGED
@@ -1,6 +1,6 @@
1
  ---
2
  license: mit
3
- base_model: neuralmind/bert-large-portuguese-cased
4
  tags:
5
  - generated_from_trainer
6
  metrics:
@@ -17,13 +17,13 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  # content
19
 
20
- This model is a fine-tuned version of [neuralmind/bert-large-portuguese-cased](https://huggingface.co/neuralmind/bert-large-portuguese-cased) on an unknown dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.4388
23
- - Accuracy: 0.7873
24
- - F1-score: 0.7774
25
- - Recall: 0.8203
26
- - Precision: 0.7388
27
 
28
  ## Model description
29
 
@@ -54,19 +54,19 @@ The following hyperparameters were used during training:
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1-score | Recall | Precision |
56
  |:-------------:|:------:|:----:|:---------------:|:--------:|:--------:|:------:|:---------:|
57
- | 0.4937 | 0.3814 | 500 | 0.4581 | 0.7828 | 0.7866 | 0.9067 | 0.6946 |
58
- | 0.4662 | 0.7628 | 1000 | 0.4117 | 0.7965 | 0.7937 | 0.8866 | 0.7185 |
59
- | 0.4121 | 1.1442 | 1500 | 0.4316 | 0.7765 | 0.7169 | 0.6410 | 0.8133 |
60
- | 0.3436 | 1.5256 | 2000 | 0.4185 | 0.8065 | 0.7894 | 0.8211 | 0.7600 |
61
- | 0.3486 | 1.9069 | 2500 | 0.4278 | 0.8085 | 0.8073 | 0.9080 | 0.7267 |
62
- | 0.2718 | 2.2883 | 3000 | 0.4940 | 0.7896 | 0.7569 | 0.7414 | 0.7730 |
63
- | 0.2437 | 2.6697 | 3500 | 0.5886 | 0.7885 | 0.7757 | 0.8283 | 0.7295 |
64
- | 0.2367 | 3.0511 | 4000 | 0.7838 | 0.7848 | 0.7690 | 0.8114 | 0.7309 |
65
- | 0.1619 | 3.4325 | 4500 | 0.7511 | 0.7842 | 0.7593 | 0.7706 | 0.7483 |
66
- | 0.1608 | 3.8139 | 5000 | 0.7884 | 0.7813 | 0.7621 | 0.7933 | 0.7334 |
67
- | 0.1168 | 4.1953 | 5500 | 1.0872 | 0.7828 | 0.7580 | 0.7706 | 0.7459 |
68
- | 0.1025 | 4.5767 | 6000 | 1.1037 | 0.7822 | 0.7659 | 0.8069 | 0.7289 |
69
- | 0.1048 | 4.9580 | 6500 | 1.1195 | 0.7842 | 0.7663 | 0.8010 | 0.7344 |
70
 
71
 
72
  ### Framework versions
 
1
  ---
2
  license: mit
3
+ base_model: neuralmind/bert-base-portuguese-cased
4
  tags:
5
  - generated_from_trainer
6
  metrics:
 
17
 
18
  # content
19
 
20
+ This model is a fine-tuned version of [neuralmind/bert-base-portuguese-cased](https://huggingface.co/neuralmind/bert-base-portuguese-cased) on an unknown dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.7314
23
+ - Accuracy: 0.7625
24
+ - F1-score: 0.7462
25
+ - Recall: 0.8237
26
+ - Precision: 0.6821
27
 
28
  ## Model description
29
 
 
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1-score | Recall | Precision |
56
  |:-------------:|:------:|:----:|:---------------:|:--------:|:--------:|:------:|:---------:|
57
+ | 0.5117 | 0.3814 | 500 | 0.4886 | 0.7659 | 0.7709 | 0.8595 | 0.6988 |
58
+ | 0.4755 | 0.7628 | 1000 | 0.4602 | 0.7584 | 0.7561 | 0.8170 | 0.7036 |
59
+ | 0.4107 | 1.1442 | 1500 | 0.5348 | 0.7730 | 0.7774 | 0.8651 | 0.7059 |
60
+ | 0.3685 | 1.5256 | 2000 | 0.4585 | 0.7728 | 0.7755 | 0.8563 | 0.7085 |
61
+ | 0.3652 | 1.9069 | 2500 | 0.4497 | 0.7802 | 0.7733 | 0.8182 | 0.7331 |
62
+ | 0.2919 | 2.2883 | 3000 | 0.5390 | 0.7659 | 0.7561 | 0.7920 | 0.7233 |
63
+ | 0.2614 | 2.6697 | 3500 | 0.5387 | 0.7636 | 0.7647 | 0.8382 | 0.7030 |
64
+ | 0.2518 | 3.0511 | 4000 | 0.6425 | 0.7679 | 0.7411 | 0.7252 | 0.7578 |
65
+ | 0.1791 | 3.4325 | 4500 | 0.6974 | 0.7682 | 0.7478 | 0.7502 | 0.7455 |
66
+ | 0.1803 | 3.8139 | 5000 | 0.6828 | 0.7831 | 0.7744 | 0.8126 | 0.7396 |
67
+ | 0.1531 | 4.1953 | 5500 | 0.8737 | 0.7690 | 0.7439 | 0.7320 | 0.7561 |
68
+ | 0.1267 | 4.5767 | 6000 | 0.9225 | 0.7730 | 0.7555 | 0.7651 | 0.7460 |
69
+ | 0.1344 | 4.9580 | 6500 | 0.9057 | 0.7753 | 0.7573 | 0.7651 | 0.7497 |
70
 
71
 
72
  ### Framework versions
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "neuralmind/bert-large-portuguese-cased",
3
  "architectures": [
4
  "BertForSequenceClassification"
5
  ],
@@ -8,14 +8,14 @@
8
  "directionality": "bidi",
9
  "hidden_act": "gelu",
10
  "hidden_dropout_prob": 0.1,
11
- "hidden_size": 1024,
12
  "initializer_range": 0.02,
13
- "intermediate_size": 4096,
14
  "layer_norm_eps": 1e-12,
15
  "max_position_embeddings": 512,
16
  "model_type": "bert",
17
- "num_attention_heads": 16,
18
- "num_hidden_layers": 24,
19
  "output_past": true,
20
  "pad_token_id": 0,
21
  "pooler_fc_size": 768,
 
1
  {
2
+ "_name_or_path": "neuralmind/bert-base-portuguese-cased",
3
  "architectures": [
4
  "BertForSequenceClassification"
5
  ],
 
8
  "directionality": "bidi",
9
  "hidden_act": "gelu",
10
  "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
  "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
  "layer_norm_eps": 1e-12,
15
  "max_position_embeddings": 512,
16
  "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
  "output_past": true,
20
  "pad_token_id": 0,
21
  "pooler_fc_size": 768,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ace2241fb1432463d8215b4e57b73e0636d3144da8804612a22f8ca9997b5084
3
- size 1337640872
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b80012ee25e3913fd5fa531e5edfe7466dd0d0118b661476fb24b668c0d11980
3
+ size 435722224
runs/Jul20_19-12-56_85df0c1db32e/events.out.tfevents.1721515789.85df0c1db32e.2808.1 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:291ef320a7989d238ed27a63c85b2db1a630570f16be612e66a1b3c11a412c9c
3
- size 1044
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a108e672957bc87006d1e4ea2838129c8a3e40651571599f0aff73943dbb603
3
+ size 1522
runs/Jul20_23-10-58_85df0c1db32e/events.out.tfevents.1721517059.85df0c1db32e.61311.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fb559f00e82c8a3a6a2f3f63ec824754ba617b42bb9527368484d5972d3f39b
3
+ size 14367
runs/Jul20_23-10-58_85df0c1db32e/events.out.tfevents.1721521186.85df0c1db32e.61311.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fed8f386e8db1572c10f3e507c6fd6c1cff81674c4927fc921d93f13dcf64ff6
3
+ size 1044
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:414f32a7ed75f4cbaef262c4f4230b4e4d23de347b8b18a42be54bca9579baf3
3
  size 5112
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79f538af0e34deb2ccfd88dba0d2e783b7ddd8e198dfa37a03c03480a41b2d83
3
  size 5112