<!-- 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. -->
BERiT_2000_ls_1
This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 6.7791
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- label_smoothing_factor: 0.1
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
6.9212 | 0.19 | 500 | 6.8322 |
6.8009 | 0.39 | 1000 | 6.8173 |
6.8004 | 0.58 | 1500 | 6.7916 |
6.7725 | 0.77 | 2000 | 6.8005 |
6.787 | 0.97 | 2500 | 6.8066 |
6.7808 | 1.16 | 3000 | 6.7838 |
6.7757 | 1.36 | 3500 | 6.7726 |
6.7847 | 1.55 | 4000 | 6.7584 |
6.7874 | 1.74 | 4500 | 6.7809 |
6.769 | 1.94 | 5000 | 6.7715 |
6.7845 | 2.13 | 5500 | 6.8000 |
6.8052 | 2.32 | 6000 | 6.7737 |
6.7496 | 2.52 | 6500 | 6.7795 |
6.7877 | 2.71 | 7000 | 6.7669 |
6.7759 | 2.9 | 7500 | 6.7791 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.2