<!-- 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-tiny-wt-48-emotion-emb-comp
This model is a fine-tuned version of gokuls/model_v1_complete_training_wt_init_48_tiny_emb_comp on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.3216
- Accuracy: 0.8885
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 |
|---|---|---|---|---|
| 1.3139 | 1.0 | 250 | 0.8529 | 0.727 |
| 0.6339 | 2.0 | 500 | 0.4745 | 0.848 |
| 0.4011 | 3.0 | 750 | 0.3605 | 0.875 |
| 0.2998 | 4.0 | 1000 | 0.3326 | 0.885 |
| 0.25 | 5.0 | 1250 | 0.3346 | 0.8815 |
| 0.2177 | 6.0 | 1500 | 0.3216 | 0.8885 |
| 0.1928 | 7.0 | 1750 | 0.3214 | 0.8885 |
| 0.1747 | 8.0 | 2000 | 0.3178 | 0.8875 |
| 0.1581 | 9.0 | 2250 | 0.3291 | 0.885 |
| 0.1404 | 10.0 | 2500 | 0.3260 | 0.887 |
Framework versions
- Transformers 4.31.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.1
- Tokenizers 0.13.3
