bert-base-uncased_peft
This model is a fine-tuned version of bert-base-uncased on the massive dataset. It achieves the following results on the evaluation set:
- Loss: 0.5427
- Accuracy: 0.8739
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: 0.0002
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
1.9899 | 1.39 | 500 | 0.8556 | 0.7993 |
0.6821 | 2.78 | 1000 | 0.6157 | 0.8544 |
0.4306 | 4.17 | 1500 | 0.5656 | 0.8578 |
0.2885 | 5.56 | 2000 | 0.5595 | 0.8711 |
0.2203 | 6.94 | 2500 | 0.6113 | 0.8815 |
0.1599 | 8.33 | 3000 | 0.6234 | 0.8765 |
0.1223 | 9.72 | 3500 | 0.6867 | 0.8726 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.14.0
- Downloads last month
- 0
Unable to determine this model's library. Check the
docs
.