Sans1807/content
This model is a fine-tuned version of Sans1807/content on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0944
- Cer: 0.0078
- Wer: 0.0406
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 25
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer | Wer |
|---|---|---|---|---|---|
| 0.0005 | 5.0761 | 1000 | 0.0714 | 0.0119 | 0.0584 |
| 0.0 | 10.1523 | 2000 | 0.0821 | 0.0115 | 0.0533 |
| 0.003 | 15.2284 | 3000 | 0.0877 | 0.0091 | 0.0381 |
| 0.0157 | 20.3046 | 4000 | 0.0944 | 0.0078 | 0.0406 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.0
- Datasets 2.17.0
- Tokenizers 0.19.1
- Downloads last month
- 47
Inference API (serverless) does not yet support transformers models for this pipeline type.