metadata
base_model: Samuael/geez_t5-15k
tags:
- generated_from_trainer
metrics:
- wer
- bleu
model-index:
- name: geez_t5-15k
results: []
geez_t5-15k
This model is a fine-tuned version of Samuael/geez_t5-15k on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.1428
- Wer: 0.2368
- Cer: 0.0611
- Bleu: 59.1856
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: 0.0001
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer | Bleu |
|---|---|---|---|---|---|---|
| 0.1721 | 1.0 | 1421 | 1.0128 | 0.2445 | 0.0640 | 58.5313 |
| 0.3157 | 2.0 | 2842 | 0.9088 | 0.2409 | 0.0625 | 59.1274 |
| 0.1958 | 3.0 | 4263 | 0.9950 | 0.2474 | 0.0617 | 58.7017 |
| 0.2275 | 4.0 | 5684 | 1.0075 | 0.2522 | 0.0638 | 58.2146 |
| 0.1139 | 5.0 | 7105 | 1.0297 | 0.2411 | 0.0607 | 59.7241 |
| 0.1243 | 6.0 | 8526 | 1.0343 | 0.2360 | 0.0608 | 59.9537 |
| 0.18 | 7.0 | 9947 | 1.0373 | 0.2423 | 0.0617 | 59.0538 |
| 0.1281 | 8.0 | 11368 | 1.1003 | 0.2344 | 0.0605 | 59.8118 |
| 0.1276 | 9.0 | 12789 | 1.1061 | 0.2385 | 0.0611 | 59.1765 |
| 0.1128 | 10.0 | 14210 | 1.1428 | 0.2368 | 0.0611 | 59.1856 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2