Florence-2-FT-JP-OCR2
This model is a fine-tuned version of microsoft/Florence-2-base-ft on EtashGuha/JapaneseDocQA dataset. It achieves the following results on the evaluation set:
- Loss: 2.7551
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-06
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 14.8441 | 0.5563 | 100 | 3.2737 |
| 14.1581 | 1.1168 | 200 | 3.1344 |
| 13.4486 | 1.6732 | 300 | 3.0362 |
| 13.287 | 2.2337 | 400 | 2.9634 |
| 12.9018 | 2.7900 | 500 | 2.9081 |
| 12.6016 | 3.3505 | 600 | 2.8683 |
| 12.5607 | 3.9068 | 700 | 2.8310 |
| 12.4259 | 4.4673 | 800 | 2.8060 |
| 12.2114 | 5.0278 | 900 | 2.7858 |
| 12.1777 | 5.5841 | 1000 | 2.7680 |
| 12.01 | 6.1446 | 1100 | 2.7604 |
| 12.0395 | 6.7010 | 1200 | 2.7551 |
Framework versions
- Transformers 4.48.0.dev0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0