<!-- 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. -->
sa_bert_12_layer_modified_complete_training_96
This model is a fine-tuned version of gokuls/sa_bert_12_layer_modified_complete_training_72_v2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.5432
- Accuracy: 0.5446
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: 1e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10000
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
2.9291 | 0.05 | 10000 | 2.7772 | 0.5161 |
2.8002 | 0.11 | 20000 | 2.7374 | 0.5210 |
2.6962 | 0.16 | 30000 | 2.6995 | 0.5256 |
2.6628 | 0.22 | 40000 | 2.6681 | 0.5296 |
2.7735 | 0.27 | 50000 | 2.6399 | 0.5332 |
2.6821 | 0.33 | 60000 | 2.6079 | 0.5366 |
2.6073 | 0.38 | 70000 | 2.5830 | 0.5397 |
2.6742 | 0.44 | 80000 | 2.5624 | 0.5422 |
2.8162 | 0.49 | 90000 | 2.5432 | 0.5446 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.1
- Tokenizers 0.13.3