metadata
library_name: transformers
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: base_sami_22k_cont_pt_ftpseudo_wr25esp5
results: []
base_sami_22k_cont_pt_ftpseudo_wr25esp5
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1150.1907
- Wer: 1.0
- Cer: 1.0
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: 0.0005
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.25
- num_epochs: 20.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 4573.469 | 1.0 | 1080 | 1150.0875 | 1.0 | 1.0 |
| 3561.5266 | 2.0 | 2160 | 1153.8236 | 1.0 | 1.0 |
| 3604.4681 | 3.0 | 3240 | 1157.7269 | 1.0 | 1.0 |
| 3641.2648 | 4.0 | 4320 | 1146.7615 | 1.0 | 1.0 |
| 3689.1912 | 5.0 | 5400 | 1146.6725 | 1.0 | 1.0 |
| 3582.5692 | 6.0 | 6480 | 1147.8068 | 1.0 | 1.0 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1
- Datasets 3.2.0
- Tokenizers 0.21.0