<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
Nso-En
This model is a fine-tuned version of kabelomalapane/nso_en_ukuxhumana_model on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.3144
- Bleu: 24.4184
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: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu |
|---|---|---|---|---|
| No log | 1.0 | 14 | 4.1292 | 11.2917 |
| No log | 2.0 | 28 | 3.8159 | 15.9321 |
| No log | 3.0 | 42 | 3.6617 | 19.7177 |
| No log | 4.0 | 56 | 3.5394 | 21.9400 |
| No log | 5.0 | 70 | 3.4525 | 23.8702 |
| No log | 6.0 | 84 | 3.3993 | 24.2223 |
| No log | 7.0 | 98 | 3.3594 | 24.7056 |
| No log | 8.0 | 112 | 3.3345 | 23.9469 |
| No log | 9.0 | 126 | 3.3183 | 24.1888 |
| No log | 10.0 | 140 | 3.3144 | 24.4184 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1
