trocr-finetuned
This model is a fine-tuned version of microsoft/trocr-small-handwritten on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 18.4189
- Cer: 2.3451
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-07
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 18.267 | 1.0 | 9 | 18.4702 | 2.3908 |
| 18.5157 | 2.0 | 18 | 18.4523 | 2.3099 |
| 17.8628 | 3.0 | 27 | 18.4189 | 2.3451 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.5.0+cu121
- Tokenizers 0.19.1
- Downloads last month
- 8
Inference API (serverless) does not yet support transformers models for this pipeline type.
Model tree for fjfur10/trocr-finetuned
Base model
microsoft/trocr-small-handwritten