README.md

metadata

license: cc-by-nc-4.0
base_model: facebook/nllb-200-distilled-600M
tags:
  - generated_from_trainer
datasets:
  - nusatranslation_mt
metrics:
  - sacrebleu
model-index:
  - name: bbc-to-ind-nmt-v2
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: nusatranslation_mt
          type: nusatranslation_mt
          config: nusatranslation_mt_btk_ind_source
          split: test
          args: nusatranslation_mt_btk_ind_source
        metrics:
          - name: Sacrebleu
            type: sacrebleu
            value: 37.5332

bbc-to-ind-nmt-v2

This model is a fine-tuned version of facebook/nllb-200-distilled-600M on the nusatranslation_mt dataset. It achieves the following results on the evaluation set:

• Loss: 1.1380
• Sacrebleu: 37.5332
• Gen Len: 37.17

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: 8
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Sacrebleu	Gen Len
4.6139	1.0	825	1.2915	32.7445	37.5685
1.1761	2.0	1650	1.1701	35.9991	37.3645
1.0383	3.0	2475	1.1361	36.9321	37.035
0.9094	4.0	3300	1.1333	37.4774	37.039
0.8158	5.0	4125	1.1380	37.5332	37.17

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.13.1
• Tokenizers 0.19.1