ditmodel
This model was fintuned on DiT model for document classification on custom dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1482
- Accuracy: 0.9523
- Weighted f1: 0.9524
- Micro f1: 0.9523
- Macro f1: 0.9505
- Weighted recall: 0.9523
- Micro recall: 0.9523
- Macro recall: 0.9523
- Weighted precision: 0.9544
- Micro precision: 0.9523
- Macro precision: 0.9506
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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Weighted f1 | Micro f1 | Macro f1 | Weighted recall | Micro recall | Macro recall | Weighted precision | Micro precision | Macro precision |
|---|
| 0.2337 | 1.0 | 78 | 0.2668 | 0.9087 | 0.9098 | 0.9087 | 0.9058 | 0.9087 | 0.9087 | 0.9040 | 0.9229 | 0.9087 | 0.9220 |
| 0.1711 | 2.0 | 156 | 0.1820 | 0.9376 | 0.9380 | 0.9376 | 0.9331 | 0.9376 | 0.9376 | 0.9403 | 0.9416 | 0.9376 | 0.9292 |
| 0.1297 | 3.0 | 234 | 0.1482 | 0.9523 | 0.9524 | 0.9523 | 0.9505 | 0.9523 | 0.9523 | 0.9523 | 0.9544 | 0.9523 | 0.9506 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.6.1
- Tokenizers 0.15.1
