arun3936/ICFOSS_Malayalam_English_Translator

nllb-200-3.3B-Malayalam_English_Translationt_nllb6

This model is a fine-tuned version of facebook/nllb-200-3.3B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0031
• Bleu: 37.4644
• Rouge: {'rouge1': 0.6947858221991348, 'rouge2': 0.47528501248267296, 'rougeL': 0.643592904253675, 'rougeLsum': 0.6438336053077185}
• Chrf: {'score': 63.562323751931785, '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: 8
• eval_batch_size: 8
• 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
1.1299	1.0	9400	1.0473	35.4794	{'rouge1': 0.6827076206592405, 'rouge2': 0.4567713815837643, 'rougeL': 0.6303031579761407, 'rougeLsum': 0.6303637744842896}	{'score': 62.07772367684291, 'char_order': 6, 'word_order': 0, 'beta': 2}
1.0391	2.0	18800	1.0172	36.5551	{'rouge1': 0.6898802619220783, 'rouge2': 0.4678566080033477, 'rougeL': 0.6376152634193879, 'rougeLsum': 0.6378050818770977}	{'score': 62.79493404105809, 'char_order': 6, 'word_order': 0, 'beta': 2}
0.9772	3.0	28200	1.0047	37.1999	{'rouge1': 0.6940761673780116, 'rouge2': 0.4729467289482048, 'rougeL': 0.6422221741064402, 'rougeLsum': 0.6423854506325695}	{'score': 63.383659426629755, 'char_order': 6, 'word_order': 0, 'beta': 2}
0.9322	4.0	37600	1.0021	37.3505	{'rouge1': 0.6946177869994575, 'rouge2': 0.47460537713160267, 'rougeL': 0.643360432984222, 'rougeLsum': 0.6434552650502989}	{'score': 63.44418689943615, 'char_order': 6, 'word_order': 0, 'beta': 2}
0.9109	5.0	47000	1.0031	37.4644	{'rouge1': 0.6947858221991348, 'rouge2': 0.47528501248267296, 'rougeL': 0.643592904253675, 'rougeLsum': 0.6438336053077185}	{'score': 63.562323751931785, 'char_order': 6, 'word_order': 0, 'beta': 2}

Framework versions

• PEFT 0.7.2.dev0
• Transformers 4.36.1
• Pytorch 2.0.1+cu117
• Datasets 2.13.1
• Tokenizers 0.15.0