metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: DEREXP
results: []
DEREXP
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.5797
- Mse: 3.5797
- Mae: 1.4414
- R2: 0.3526
- Accuracy: 0.2268
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Mse | Mae | R2 | Accuracy |
|---|---|---|---|---|---|---|---|
| 14.19 | 0.08 | 500 | 4.6174 | 4.6174 | 1.6744 | 0.1649 | 0.198 |
| 4.527 | 0.16 | 1000 | 3.9019 | 3.9019 | 1.5164 | 0.2943 | 0.2192 |
| 4.3036 | 0.24 | 1500 | 5.3501 | 5.3501 | 1.8130 | 0.0324 | 0.1736 |
| 4.0923 | 0.32 | 2000 | 3.8948 | 3.8948 | 1.5150 | 0.2956 | 0.2142 |
| 4.0042 | 0.4 | 2500 | 3.7648 | 3.7648 | 1.4905 | 0.3191 | 0.2162 |
| 3.8685 | 0.48 | 3000 | 3.7741 | 3.7741 | 1.4908 | 0.3174 | 0.2152 |
| 3.8928 | 0.56 | 3500 | 3.7122 | 3.7122 | 1.4738 | 0.3286 | 0.214 |
| 3.8193 | 0.64 | 4000 | 3.7020 | 3.7020 | 1.4727 | 0.3304 | 0.2182 |
| 3.6929 | 0.72 | 4500 | 3.6419 | 3.6419 | 1.4575 | 0.3413 | 0.2266 |
| 3.7974 | 0.8 | 5000 | 3.6995 | 3.6995 | 1.4656 | 0.3309 | 0.2202 |
| 3.7752 | 0.88 | 5500 | 3.6344 | 3.6344 | 1.4559 | 0.3427 | 0.2276 |
| 3.6254 | 0.96 | 6000 | 3.5797 | 3.5797 | 1.4414 | 0.3526 | 0.2268 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.12.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1