donut_synDB_945_linear
This model is a fine-tuned version of naver-clova-ix/donut-base on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.1215
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: 0.0001
- train_batch_size: 5
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 10
- total_train_batch_size: 50
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 20
- num_epochs: 7
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.9306 | 1.0 | 20 | 0.7244 |
0.7015 | 1.5 | 30 | 0.3480 |
0.3645 | 2.0 | 40 | 0.2467 |
0.212 | 2.5 | 50 | 0.1763 |
0.1488 | 3.0 | 60 | 0.1523 |
0.1022 | 3.5 | 70 | 0.1261 |
0.1014 | 4.0 | 80 | 0.1258 |
0.0732 | 4.5 | 90 | 0.1308 |
0.0663 | 5.0 | 100 | 0.1239 |
0.0528 | 5.5 | 110 | 0.1241 |
0.0569 | 6.0 | 120 | 0.1226 |
0.0485 | 6.5 | 130 | 0.1219 |
0.0484 | 7.0 | 140 | 0.1215 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 14
Unable to determine this model’s pipeline type. Check the
docs
.