nllb-200-1.3B-ICFOSS-Hindi_Malayalam_Translator

This model is a fine-tuned version of facebook/nllb-200-1.3B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5883
• Bleu: 32.3133
• Rouge: {'rouge1': 0.4065168353303946, 'rouge2': 0.2762287202150305, 'rougeL': 0.3947284265080875, 'rougeLsum': 0.3952044100349186}
• Chrf: {'score': 69.00192526889488, 'char_order': 6, 'word_order': 0, 'beta': 2}

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: 0.0002
• 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: cosine
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Rouge	Chrf
0.7199	1.0	4698	0.6215	29.8169	{'rouge1': 0.405762448982788, 'rouge2': 0.2722973168668976, 'rougeL': 0.3920717141056125, 'rougeLsum': 0.39226265285587325}	{'score': 67.671645233609, 'char_order': 6, 'word_order': 0, 'beta': 2}
0.6269	2.0	9396	0.5960	31.5169	{'rouge1': 0.4074341383663418, 'rouge2': 0.2754310444010575, 'rougeL': 0.3950826699767377, 'rougeLsum': 0.3954069337543914}	{'score': 68.50668767972792, 'char_order': 6, 'word_order': 0, 'beta': 2}
0.5962	3.0	14094	0.5891	32.4131	{'rouge1': 0.4065168353303946, 'rouge2': 0.2774409260882534, 'rougeL': 0.3949096085748628, 'rougeLsum': 0.39546625690693493}	{'score': 68.94783702655978, 'char_order': 6, 'word_order': 0, 'beta': 2}
0.5855	4.0	18792	0.5882	32.4648	{'rouge1': 0.4065168353303946, 'rouge2': 0.2762287202150305, 'rougeL': 0.3947284265080875, 'rougeLsum': 0.3952044100349186}	{'score': 69.10087499970177, 'char_order': 6, 'word_order': 0, 'beta': 2}
0.5835	5.0	23490	0.5883	32.3133	{'rouge1': 0.4065168353303946, 'rouge2': 0.2762287202150305, 'rougeL': 0.3947284265080875, 'rougeLsum': 0.3952044100349186}	{'score': 69.00192526889488, 'char_order': 6, 'word_order': 0, 'beta': 2}

Framework versions

• PEFT 0.10.0
• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1