nllb-en-id-ccmatrix

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

• Loss: 0.4791
• Bleu: 65.9837

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 4000
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
0.606	1.0	28125	0.5249	64.1268
0.4943	2.0	56250	0.5043	64.7892
0.467	3.0	84375	0.4945	65.2331
0.4487	4.0	112500	0.4887	65.5512
0.4349	5.0	140625	0.4843	65.6806
0.4242	6.0	168750	0.4822	65.7774
0.416	7.0	196875	0.4801	65.8541
0.4098	8.0	225000	0.4800	65.9652
0.4052	9.0	253125	0.4788	65.9701
0.4023	10.0	281250	0.4791	65.9837

Framework versions

• Transformers 4.26.1
• Pytorch 2.0.0
• Datasets 2.10.1
• Tokenizers 0.11.0