<!-- 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. -->
hbertv1-wt-frz-48-emotion
This model is a fine-tuned version of gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48_frz on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.3306
- Accuracy: 0.9195
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: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 33
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.8962 | 1.0 | 250 | 0.3587 | 0.872 |
| 0.3328 | 2.0 | 500 | 0.3154 | 0.889 |
| 0.2269 | 3.0 | 750 | 0.2463 | 0.913 |
| 0.1687 | 4.0 | 1000 | 0.3033 | 0.912 |
| 0.1319 | 5.0 | 1250 | 0.2559 | 0.9105 |
| 0.1091 | 6.0 | 1500 | 0.2657 | 0.913 |
| 0.0809 | 7.0 | 1750 | 0.3015 | 0.913 |
| 0.0686 | 8.0 | 2000 | 0.3306 | 0.9195 |
| 0.0498 | 9.0 | 2250 | 0.3532 | 0.9195 |
| 0.0389 | 10.0 | 2500 | 0.3960 | 0.9175 |
Framework versions
- Transformers 4.31.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.1
- Tokenizers 0.13.3
