<!-- 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_L12_H64_A2_emotion
This model is a fine-tuned version of gokuls/HBERTv1_48_L12_H64_A2 on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.5281
- Accuracy: 0.8265
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 |
---|---|---|---|---|
1.606 | 1.0 | 250 | 1.4528 | 0.4995 |
1.32 | 2.0 | 500 | 1.1600 | 0.5915 |
1.0418 | 3.0 | 750 | 0.9467 | 0.6765 |
0.8368 | 4.0 | 1000 | 0.7801 | 0.7415 |
0.6914 | 5.0 | 1250 | 0.6631 | 0.783 |
0.5831 | 6.0 | 1500 | 0.5996 | 0.809 |
0.5242 | 7.0 | 1750 | 0.5723 | 0.81 |
0.4816 | 8.0 | 2000 | 0.5426 | 0.819 |
0.4544 | 9.0 | 2250 | 0.5318 | 0.824 |
0.4276 | 10.0 | 2500 | 0.5281 | 0.8265 |
Framework versions
- Transformers 4.34.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.14.5
- Tokenizers 0.14.0