<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
clickbait_binary_detection
This model is a fine-tuned version of roberta-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4630
- Macro F1: 0.9155
- Micro F1: 0.9215
- Accuracy: 0.9215
Performance on test set:
-
Accuracy: 0.9257990867579908
-
F1 score: 0.9199282431058413
-
Precision: 0.9233793490724882
-
Recall : 0.9168756883647268
-
Matthews Correlation Coefficient: 0.8402298675576902
-
Precision of each class: [0.931899 0.91485969]
-
Recall of each class: [0.95152505 0.88222632]
-
F1 score of each class: [0.94160977 0.89824671]
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-06
- train_batch_size: 6
- eval_batch_size: 10
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 12
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Macro F1 | Micro F1 | Accuracy |
|---|---|---|---|---|---|---|
| 0.2296 | 1.0 | 3650 | 0.2236 | 0.9105 | 0.9183 | 0.9183 |
| 0.228 | 2.0 | 7301 | 0.2708 | 0.9115 | 0.9192 | 0.9192 |
| 0.2075 | 3.0 | 10951 | 0.3141 | 0.9164 | 0.9224 | 0.9224 |
| 0.1881 | 4.0 | 14602 | 0.3211 | 0.9143 | 0.9201 | 0.9201 |
| 0.18 | 5.0 | 18252 | 0.3852 | 0.9130 | 0.9188 | 0.9188 |
| 0.1818 | 6.0 | 21903 | 0.3784 | 0.9110 | 0.9174 | 0.9174 |
| 0.1495 | 7.0 | 25553 | 0.4606 | 0.9106 | 0.9156 | 0.9156 |
| 0.1453 | 8.0 | 29204 | 0.4630 | 0.9155 | 0.9215 | 0.9215 |
Framework versions
- Transformers 4.27.1
- Pytorch 2.0.1+cu118
- Datasets 2.9.0
- Tokenizers 0.13.3
