donut-base-sroie-metrics-combined-new-fixed-version
This model is a fine-tuned version of naver-clova-ix/donut-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4187
- Bleu: 0.0659
- Precisions: [0.8259958071278826, 0.7547619047619047, 0.7107438016528925, 0.6601307189542484]
- Brevity Penalty: 0.0895
- Length Ratio: 0.2930
- Translation Length: 477
- Reference Length: 1628
- Cer: 0.7531
- Wer: 0.8260
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 2
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Bleu | Precisions | Brevity Penalty | Length Ratio | Translation Length | Reference Length | Cer | Wer |
---|---|---|---|---|---|---|---|---|---|---|---|
0.9081 | 1.0 | 253 | 0.5714 | 0.0513 | [0.7483731019522777, 0.6584158415841584, 0.6138328530259366, 0.5724137931034483] | 0.0795 | 0.2832 | 461 | 1628 | 0.7796 | 0.8572 |
0.3952 | 2.0 | 506 | 0.4489 | 0.0579 | [0.7913978494623656, 0.7254901960784313, 0.6809116809116809, 0.6360544217687075] | 0.0820 | 0.2856 | 465 | 1628 | 0.7632 | 0.8394 |
0.3077 | 3.0 | 759 | 0.4266 | 0.0666 | [0.8218029350104822, 0.7642857142857142, 0.721763085399449, 0.673202614379085] | 0.0895 | 0.2930 | 477 | 1628 | 0.7556 | 0.8273 |
0.2307 | 4.0 | 1012 | 0.4187 | 0.0659 | [0.8259958071278826, 0.7547619047619047, 0.7107438016528925, 0.6601307189542484] | 0.0895 | 0.2930 | 477 | 1628 | 0.7531 | 0.8260 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.1.0
- Datasets 2.18.0
- Tokenizers 0.19.1
- Downloads last month
- 0
Unable to determine this model’s pipeline type. Check the
docs
.