<!-- 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. -->
hbertv2-emotion-48-emb-comp-gelu
This model is a fine-tuned version of gokuls/bert_12_layer_model_v2_complete_training_new_emb_compress_48_gelu on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.6500
- Accuracy: 0.8125
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 |
|---|---|---|---|---|
| 1.583 | 1.0 | 250 | 1.5406 | 0.3875 |
| 1.3735 | 2.0 | 500 | 1.2460 | 0.5365 |
| 1.1696 | 3.0 | 750 | 1.1673 | 0.556 |
| 1.0567 | 4.0 | 1000 | 1.0862 | 0.574 |
| 0.8667 | 5.0 | 1250 | 0.8843 | 0.686 |
| 0.6994 | 6.0 | 1500 | 0.8536 | 0.698 |
| 0.5608 | 7.0 | 1750 | 0.7322 | 0.773 |
| 0.4448 | 8.0 | 2000 | 0.6712 | 0.8045 |
| 0.3793 | 9.0 | 2250 | 0.6298 | 0.8095 |
| 0.335 | 10.0 | 2500 | 0.6500 | 0.8125 |
Framework versions
- Transformers 4.31.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.1
- Tokenizers 0.13.3
