nllb-200-distilled-1.3B-ICFOSS-Tamil_Malayalam_Translation2

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

• Loss: 1.1621
• Bleu: 17.6076
• Rouge: {'rouge1': 0.23715147352380064, 'rouge2': 0.12071739418595513, 'rougeL': 0.2345381430444835, 'rougeLsum': 0.23453506374330857}
• Chrf: {'score': 53.10170184949962, '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
1.3104	1.0	3200	1.1873	16.9649	{'rouge1': 0.2373291149186338, 'rouge2': 0.12055406484331477, 'rougeL': 0.2349602512054163, 'rougeLsum': 0.2347613356578951}	{'score': 52.63432691341799, 'char_order': 6, 'word_order': 0, 'beta': 2}
1.2392	2.0	6400	1.1684	17.4803	{'rouge1': 0.23688544501541525, 'rouge2': 0.12054604364691682, 'rougeL': 0.23435163398707426, 'rougeLsum': 0.23426003897018}	{'score': 53.023399634389435, 'char_order': 6, 'word_order': 0, 'beta': 2}
1.2206	3.0	9600	1.1636	17.5799	{'rouge1': 0.2378202693811884, 'rouge2': 0.12098156404604737, 'rougeL': 0.2353625364071846, 'rougeLsum': 0.2352423617840227}	{'score': 53.09262212159299, 'char_order': 6, 'word_order': 0, 'beta': 2}
1.2165	4.0	12800	1.1620	17.5801	{'rouge1': 0.23733760623984934, 'rouge2': 0.12071579562905231, 'rougeL': 0.23472198687403475, 'rougeLsum': 0.23465210830971256}	{'score': 53.058606903092645, 'char_order': 6, 'word_order': 0, 'beta': 2}
1.214	5.0	16000	1.1621	17.6076	{'rouge1': 0.23715147352380064, 'rouge2': 0.12071739418595513, 'rougeL': 0.2345381430444835, 'rougeLsum': 0.23453506374330857}	{'score': 53.10170184949962, 'char_order': 6, 'word_order': 0, 'beta': 2}

Framework versions

• PEFT 0.10.0
• Transformers 4.39.3
• Pytorch 2.1.0+cu121
• Datasets 2.19.0
• Tokenizers 0.15.0