Hugging Face's logo

callmesan
/
twitter-roberta-large-hate-latest-profanity-mr

Text Classification
Transformers
Safetensors
roberta
Generated from Trainer
Inference Endpoints
Model card Files Community
callmesan commited on
Commit
79c8e6b
•
1 Parent(s): 3903464

End of training

Browse files
Files changed (2) hide show
  1. README.md +15 -15
  2. model.safetensors +1 -1
README.md CHANGED
@@ -21,7 +21,7 @@ should probably proofread and complete it, then remove this comment. -->
21
 
22
  This model is a fine-tuned version of [cardiffnlp/twitter-roberta-large-hate-latest](https://huggingface.co/cardiffnlp/twitter-roberta-large-hate-latest) on the None dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.3169
25
  - Accuracy: 0.9035
26
  - Precision: 0.4517
27
  - Recall: 0.5
@@ -45,29 +45,29 @@ More information needed
45
 
46
  The following hyperparameters were used during training:
47
  - learning_rate: 5e-05
48
- - train_batch_size: 16
49
  - eval_batch_size: 128
50
  - seed: 42
51
  - gradient_accumulation_steps: 2
52
- - total_train_batch_size: 32
53
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
54
  - lr_scheduler_type: linear
55
  - num_epochs: 10
56
 
57
  ### Training results
58
 
59
- | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
60
- |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
61
- | 0.2867 | 1.0 | 61 | 0.3645 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
62
- | 0.3329 | 2.0 | 122 | 0.4653 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
63
- | 0.3481 | 3.0 | 183 | 0.3754 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
64
- | 0.3655 | 4.0 | 244 | 0.3711 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
65
- | 0.2946 | 5.0 | 305 | 0.3984 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
66
- | 0.2919 | 6.0 | 366 | 0.3844 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
67
- | 0.3337 | 7.0 | 427 | 0.3738 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
68
- | 0.2662 | 8.0 | 488 | 0.3753 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
69
- | 0.2621 | 9.0 | 549 | 0.3906 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
70
- | 0.3298 | 10.0 | 610 | 0.3755 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
71
 
72
 
73
  ### Framework versions
 
21
 
22
  This model is a fine-tuned version of [cardiffnlp/twitter-roberta-large-hate-latest](https://huggingface.co/cardiffnlp/twitter-roberta-large-hate-latest) on the None dataset.
23
  It achieves the following results on the evaluation set:
24
+ - Loss: 0.3265
25
  - Accuracy: 0.9035
26
  - Precision: 0.4517
27
  - Recall: 0.5
 
45
 
46
  The following hyperparameters were used during training:
47
  - learning_rate: 5e-05
48
+ - train_batch_size: 32
49
  - eval_batch_size: 128
50
  - seed: 42
51
  - gradient_accumulation_steps: 2
52
+ - total_train_batch_size: 64
53
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
54
  - lr_scheduler_type: linear
55
  - num_epochs: 10
56
 
57
  ### Training results
58
 
59
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
60
+ |:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
61
+ | 0.3253 | 0.9836 | 30 | 0.3848 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
62
+ | 0.3423 | 2.0 | 61 | 0.3673 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
63
+ | 0.3399 | 2.9836 | 91 | 0.3823 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
64
+ | 0.3248 | 4.0 | 122 | 0.3630 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
65
+ | 0.3 | 4.9836 | 152 | 0.3922 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
66
+ | 0.3094 | 6.0 | 183 | 0.3655 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
67
+ | 0.3009 | 6.9836 | 213 | 0.3835 | 0.8819 | 0.4410 | 0.5 | 0.4686 |
68
+ | 0.2442 | 8.0 | 244 | 0.4946 | 0.7904 | 0.5976 | 0.6514 | 0.6105 |
69
+ | 0.1438 | 8.9836 | 274 | 0.3878 | 0.8699 | 0.6590 | 0.5992 | 0.6179 |
70
+ | 0.1372 | 9.8361 | 300 | 0.3606 | 0.8723 | 0.6665 | 0.6006 | 0.6207 |
71
 
72
 
73
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b1e9d7042355aea0e57224b7dd27c10bb76beb8f02100386873fd0f9d4f3d760
3
  size 1421495416
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4460cf7f9970ecf25e682e5080d765bef7550fc839ad144d1542c14af4cea75d
3
  size 1421495416