metadata
license: apache-2.0
base_model: Helsinki-NLP/opus-mt-en-ro
tags:
- generated_from_trainer
datasets:
- arrow
metrics:
- bleu
model-index:
- name: opus-mt-en-bkm-Final-37
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: arrow
type: arrow
config: default
split: train
args: default
metrics:
- name: Bleu
type: bleu
value: 0.8035
opus-mt-en-bkm-Final-37
This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-ro on the arrow dataset. It achieves the following results on the evaluation set:
- Loss: 2.1287
- Bleu: 0.8035
- Gen Len: 8.152
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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| No log | 1.0 | 364 | 2.7636 | 1.8891 | 7.3867 |
| 3.6413 | 2.0 | 728 | 2.4732 | 1.2427 | 7.7232 |
| 2.5061 | 3.0 | 1092 | 2.3356 | 0.2297 | 8.558 |
| 2.5061 | 4.0 | 1456 | 2.2540 | 0.4676 | 8.8199 |
| 2.212 | 5.0 | 1820 | 2.2023 | 0.4941 | 9.0923 |
| 2.0058 | 6.0 | 2184 | 2.1658 | 0.665 | 8.627 |
| 1.8426 | 7.0 | 2548 | 2.1516 | 0.7263 | 8.4051 |
| 1.8426 | 8.0 | 2912 | 2.1404 | 0.7423 | 8.3173 |
| 1.7227 | 9.0 | 3276 | 2.1323 | 0.7364 | 8.3373 |
| 1.6471 | 10.0 | 3640 | 2.1287 | 0.8035 | 8.152 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2