mbartLarge_mid_en-ko1
This model is a fine-tuned version of facebook/mbart-large-50-many-to-many-mmt on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.5230
- Bleu: 11.2393
- Gen Len: 16.3612
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: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 16
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 40
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 1.4343 | 1.12 | 1500 | 1.5230 | 11.2393 | 16.3612 |
| 0.9153 | 2.24 | 3000 | 1.6003 | 11.9955 | 16.2665 |
| 0.4823 | 3.37 | 4500 | 1.7483 | 11.5074 | 16.1329 |
| 0.3021 | 4.49 | 6000 | 1.8885 | 11.452 | 16.2631 |
| 0.1974 | 5.61 | 7500 | 2.0004 | 12.065 | 16.0183 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.5
- Tokenizers 0.14.1
- Downloads last month
- 7