<!-- 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_48_L6_H128_A2_emotion
This model is a fine-tuned version of gokuls/HBERTv1_48_L6_H128_A2 on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.4045
- Accuracy: 0.8725
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 |
---|---|---|---|---|
1.4612 | 1.0 | 250 | 1.2379 | 0.56 |
1.0531 | 2.0 | 500 | 0.8848 | 0.679 |
0.7465 | 3.0 | 750 | 0.6610 | 0.767 |
0.5388 | 4.0 | 1000 | 0.5209 | 0.8295 |
0.4274 | 5.0 | 1250 | 0.4551 | 0.8525 |
0.3591 | 6.0 | 1500 | 0.4346 | 0.8585 |
0.3107 | 7.0 | 1750 | 0.4174 | 0.863 |
0.2807 | 8.0 | 2000 | 0.4125 | 0.8705 |
0.2602 | 9.0 | 2250 | 0.4045 | 0.8725 |
0.241 | 10.0 | 2500 | 0.4108 | 0.872 |
Framework versions
- Transformers 4.34.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.14.5
- Tokenizers 0.14.0