<!-- 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_L10_H64_A2_emotion
This model is a fine-tuned version of gokuls/HBERTv1_48_L10_H64_A2 on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.6478
- Accuracy: 0.788
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 |
---|---|---|---|---|
1.6249 | 1.0 | 250 | 1.4994 | 0.4705 |
1.4034 | 2.0 | 500 | 1.2998 | 0.5165 |
1.2226 | 3.0 | 750 | 1.1387 | 0.5805 |
1.06 | 4.0 | 1000 | 0.9972 | 0.6415 |
0.9194 | 5.0 | 1250 | 0.8887 | 0.6965 |
0.8111 | 6.0 | 1500 | 0.7972 | 0.7395 |
0.7282 | 7.0 | 1750 | 0.7228 | 0.7575 |
0.6637 | 8.0 | 2000 | 0.6712 | 0.775 |
0.6108 | 9.0 | 2250 | 0.6478 | 0.788 |
0.585 | 10.0 | 2500 | 0.6436 | 0.7865 |
Framework versions
- Transformers 4.34.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.14.5
- Tokenizers 0.14.0