<!-- 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-emotion_48_KD_w_in
This model is a fine-tuned version of gokuls/bert_12_layer_model_v1_complete_training_new_48_KD_wt_init on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.2520
- Accuracy: 0.9245
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 |
|---|---|---|---|---|
| 0.8453 | 1.0 | 250 | 0.4759 | 0.8615 |
| 0.3814 | 2.0 | 500 | 0.3301 | 0.889 |
| 0.2766 | 3.0 | 750 | 0.3088 | 0.8975 |
| 0.2282 | 4.0 | 1000 | 0.2545 | 0.9055 |
| 0.1812 | 5.0 | 1250 | 0.2287 | 0.9135 |
| 0.1478 | 6.0 | 1500 | 0.2520 | 0.9245 |
| 0.1223 | 7.0 | 1750 | 0.2605 | 0.923 |
| 0.0982 | 8.0 | 2000 | 0.2812 | 0.916 |
| 0.0806 | 9.0 | 2250 | 0.2943 | 0.918 |
| 0.064 | 10.0 | 2500 | 0.3078 | 0.9155 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.0
- Tokenizers 0.13.3
