Hugging Face's logo

franac1
/
hate_speech

Text Classification
Transformers
Safetensors
electra
Generated from Trainer
Model card Files
xet
Community
franac1 commited on
Commit
77d6c39
·
verified ·
1 Parent(s): bc5920c

Training in progress, epoch 0

Browse files
Files changed (5) hide show
  1. README.md +24 -23
  2. config.json +2 -6
  3. model.safetensors +2 -2
  4. special_tokens_map.json +35 -5
  5. training_args.bin +1 -1
README.md CHANGED
@@ -21,12 +21,12 @@ should probably proofread and complete it, then remove this comment. -->
21
 
22
  This model is a fine-tuned version of [classla/bcms-bertic](https://huggingface.co/classla/bcms-bertic) on an unknown dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.6168
25
- - Accuracy: 0.7829
26
- - Auc Score: 0.8669
27
- - F1: 0.8083
28
- - Precision: 0.7949
29
- - Recall: 0.8221
30
 
31
  ## Model description
32
 
@@ -57,23 +57,24 @@ The following hyperparameters were used during training:
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | Auc Score | F1 | Precision | Recall |
59
  |:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:---------:|:------:|
60
- | 0.6346 | 0.1736 | 100 | 0.5729 | 0.7036 | 0.8184 | 0.7729 | 0.6738 | 0.9061 |
61
- | 0.5153 | 0.3472 | 200 | 0.5228 | 0.7522 | 0.8334 | 0.7893 | 0.7490 | 0.8343 |
62
- | 0.524 | 0.5208 | 300 | 0.6139 | 0.7239 | 0.8494 | 0.7918 | 0.6821 | 0.9436 |
63
- | 0.5021 | 0.6944 | 400 | 0.4913 | 0.7675 | 0.8467 | 0.7946 | 0.7818 | 0.8077 |
64
- | 0.4612 | 0.8681 | 500 | 0.5028 | 0.7601 | 0.8612 | 0.8038 | 0.7378 | 0.8829 |
65
- | 0.4288 | 1.0417 | 600 | 0.4997 | 0.7804 | 0.8603 | 0.8080 | 0.7872 | 0.8298 |
66
- | 0.3585 | 1.2153 | 700 | 0.5485 | 0.7700 | 0.8600 | 0.8038 | 0.7652 | 0.8464 |
67
- | 0.4089 | 1.3889 | 800 | 0.4786 | 0.7860 | 0.8626 | 0.8077 | 0.8077 | 0.8077 |
68
- | 0.3785 | 1.5625 | 900 | 0.4961 | 0.7749 | 0.8630 | 0.7832 | 0.8442 | 0.7304 |
69
- | 0.3513 | 1.7361 | 1000 | 0.5348 | 0.7780 | 0.8647 | 0.8064 | 0.7833 | 0.8309 |
70
- | 0.3674 | 1.9097 | 1100 | 0.4934 | 0.7823 | 0.8674 | 0.8007 | 0.8163 | 0.7856 |
71
- | 0.2826 | 2.0833 | 1200 | 0.5916 | 0.7792 | 0.8680 | 0.8106 | 0.7758 | 0.8486 |
72
- | 0.2445 | 2.2569 | 1300 | 0.6143 | 0.7706 | 0.8648 | 0.8058 | 0.7618 | 0.8552 |
73
- | 0.247 | 2.4306 | 1400 | 0.5840 | 0.7811 | 0.8630 | 0.8069 | 0.7923 | 0.8221 |
74
- | 0.2443 | 2.6042 | 1500 | 0.5923 | 0.7872 | 0.8676 | 0.8054 | 0.8202 | 0.7912 |
75
- | 0.2512 | 2.7778 | 1600 | 0.5986 | 0.7860 | 0.8674 | 0.8032 | 0.8227 | 0.7845 |
76
- | 0.2119 | 2.9514 | 1700 | 0.6168 | 0.7829 | 0.8669 | 0.8083 | 0.7949 | 0.8221 |
 
77
 
78
 
79
  ### Framework versions
 
21
 
22
  This model is a fine-tuned version of [classla/bcms-bertic](https://huggingface.co/classla/bcms-bertic) on an unknown dataset.
23
  It achieves the following results on the evaluation set:
24
+ - Loss: 0.5390
25
+ - Accuracy: 0.7669
26
+ - Auc Score: 0.8334
27
+ - F1: 0.6965
28
+ - Precision: 0.6768
29
+ - Recall: 0.7174
30
 
31
  ## Model description
32
 
 
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | Auc Score | F1 | Precision | Recall |
59
  |:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:---------:|:------:|
60
+ | 0.643 | 0.1616 | 100 | 0.6137 | 0.6403 | 0.7142 | 0.1180 | 0.6885 | 0.0645 |
61
+ | 0.6015 | 0.3231 | 200 | 0.5704 | 0.6861 | 0.7557 | 0.6257 | 0.5633 | 0.7035 |
62
+ | 0.5784 | 0.4847 | 300 | 0.5420 | 0.7153 | 0.7807 | 0.5667 | 0.6552 | 0.4992 |
63
+ | 0.5545 | 0.6462 | 400 | 0.5177 | 0.7394 | 0.8052 | 0.5991 | 0.7025 | 0.5223 |
64
+ | 0.5328 | 0.8078 | 500 | 0.5342 | 0.7194 | 0.8100 | 0.6729 | 0.5950 | 0.7742 |
65
+ | 0.5524 | 0.9693 | 600 | 0.5678 | 0.6976 | 0.8124 | 0.6796 | 0.5617 | 0.8602 |
66
+ | 0.4563 | 1.1309 | 700 | 0.5280 | 0.7423 | 0.8101 | 0.6596 | 0.6498 | 0.6697 |
67
+ | 0.4425 | 1.2924 | 800 | 0.5082 | 0.7549 | 0.8202 | 0.6480 | 0.6973 | 0.6052 |
68
+ | 0.4436 | 1.4540 | 900 | 0.5129 | 0.7537 | 0.8250 | 0.6356 | 0.7089 | 0.5760 |
69
+ | 0.42 | 1.6155 | 1000 | 0.5073 | 0.7635 | 0.8242 | 0.6656 | 0.7038 | 0.6313 |
70
+ | 0.426 | 1.7771 | 1100 | 0.5008 | 0.7635 | 0.8276 | 0.6816 | 0.6842 | 0.6790 |
71
+ | 0.4425 | 1.9386 | 1200 | 0.5007 | 0.7646 | 0.8337 | 0.6940 | 0.6734 | 0.7158 |
72
+ | 0.3873 | 2.1002 | 1300 | 0.5566 | 0.7646 | 0.8377 | 0.7028 | 0.6639 | 0.7465 |
73
+ | 0.306 | 2.2617 | 1400 | 0.5551 | 0.7600 | 0.8343 | 0.6939 | 0.6616 | 0.7296 |
74
+ | 0.3297 | 2.4233 | 1500 | 0.5707 | 0.7726 | 0.8309 | 0.6700 | 0.7301 | 0.6190 |
75
+ | 0.3335 | 2.5848 | 1600 | 0.5412 | 0.7686 | 0.8338 | 0.6878 | 0.6921 | 0.6836 |
76
+ | 0.3149 | 2.7464 | 1700 | 0.5438 | 0.7692 | 0.8338 | 0.6864 | 0.6956 | 0.6774 |
77
+ | 0.3577 | 2.9079 | 1800 | 0.5390 | 0.7669 | 0.8334 | 0.6965 | 0.6768 | 0.7174 |
78
 
79
 
80
  ### Framework versions
config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "architectures": [
3
- "ElectraForSequenceClassification"
4
  ],
5
  "attention_probs_dropout_prob": 0.1,
6
  "classifier_dropout": null,
@@ -12,16 +12,12 @@
12
  "intermediate_size": 3072,
13
  "layer_norm_eps": 1e-12,
14
  "max_position_embeddings": 512,
15
- "model_type": "electra",
16
  "num_attention_heads": 12,
17
  "num_hidden_layers": 12,
18
  "pad_token_id": 0,
19
  "position_embedding_type": "absolute",
20
  "problem_type": "single_label_classification",
21
- "summary_activation": "gelu",
22
- "summary_last_dropout": 0.1,
23
- "summary_type": "first",
24
- "summary_use_proj": true,
25
  "torch_dtype": "float32",
26
  "transformers_version": "4.52.4",
27
  "type_vocab_size": 2,
 
1
  {
2
  "architectures": [
3
+ "BertForSequenceClassification"
4
  ],
5
  "attention_probs_dropout_prob": 0.1,
6
  "classifier_dropout": null,
 
12
  "intermediate_size": 3072,
13
  "layer_norm_eps": 1e-12,
14
  "max_position_embeddings": 512,
15
+ "model_type": "bert",
16
  "num_attention_heads": 12,
17
  "num_hidden_layers": 12,
18
  "pad_token_id": 0,
19
  "position_embedding_type": "absolute",
20
  "problem_type": "single_label_classification",
 
 
 
 
21
  "torch_dtype": "float32",
22
  "transformers_version": "4.52.4",
23
  "type_vocab_size": 2,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:89fd178b278ec5576ce071464e05e1667d667d3107ff6d08d3f0a3af757a949c
3
- size 442499672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dedf46e7b2d8eae3eee4faf49893f6a32e9f9cf038e0502179deaedec53e531
3
+ size 442499064
special_tokens_map.json CHANGED
@@ -1,7 +1,37 @@
1
  {
2
- "cls_token": "[CLS]",
3
- "mask_token": "[MASK]",
4
- "pad_token": "[PAD]",
5
- "sep_token": "[SEP]",
6
- "unk_token": "[UNK]"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  }
 
1
  {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c04be30d77da3124f89838d95e5c5f73fe8812c3eb70775b5f59d6c5ccc1d7e9
3
  size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7dad0e36d84adbb572d1c02d802bad9a42408ce4be2b9c4ba29fae3b3465657d
3
  size 5304