nllb-200-1.3B-ICFOSS_Malayalam_Tamil_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: 0.8336
• Bleu: 30.1755
• Rouge: {'rouge1': 0.27813852813852813, 'rouge2': 0.14151205936920222, 'rougeL': 0.268193413729128, 'rougeLsum': 0.2691068851783137}
• Chrf: {'score': 66.57581227024936, '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
0.9998	1.0	3806	0.8684	28.6085	{'rouge1': 0.2776283240568955, 'rouge2': 0.14373067408781692, 'rougeL': 0.26886982065553494, 'rougeLsum': 0.2699984539270253}	{'score': 65.8388692774937, 'char_order': 6, 'word_order': 0, 'beta': 2}
0.8913	2.0	7612	0.8432	29.6027	{'rouge1': 0.27813852813852813, 'rouge2': 0.14151205936920222, 'rougeL': 0.268193413729128, 'rougeLsum': 0.2691068851783137}	{'score': 66.3019387532507, 'char_order': 6, 'word_order': 0, 'beta': 2}
0.8506	3.0	11418	0.8330	29.8285	{'rouge1': 0.27813852813852813, 'rouge2': 0.14151205936920222, 'rougeL': 0.268193413729128, 'rougeLsum': 0.2691068851783137}	{'score': 66.55262372617446, 'char_order': 6, 'word_order': 0, 'beta': 2}
0.8356	4.0	15224	0.8336	30.1355	{'rouge1': 0.27813852813852813, 'rouge2': 0.14336734693877548, 'rougeL': 0.2690772521129664, 'rougeLsum': 0.2704442383013811}	{'score': 66.59248639173376, 'char_order': 6, 'word_order': 0, 'beta': 2}
0.8321	5.0	19030	0.8336	30.1755	{'rouge1': 0.27813852813852813, 'rouge2': 0.14151205936920222, 'rougeL': 0.268193413729128, 'rougeLsum': 0.2691068851783137}	{'score': 66.57581227024936, '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