mt5-base-ICFOSS-malayalam_Hindi_Translator

This model is a fine-tuned version of google/mt5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.2179
• Bleu: 6.2035
• Rouge: {'rouge1': 0.2667970960136926, 'rouge2': 0.14574925525428614, 'rougeL': 0.26511828595423204, 'rougeLsum': 0.26501665904942706}
• Chrf: {'score': 23.454551827072866, '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
2.5515	1.0	4315	1.2874	5.8306	{'rouge1': 0.2660910934739513, 'rouge2': 0.14404792849379128, 'rougeL': 0.26384549634107013, 'rougeLsum': 0.2637751499455684}	{'score': 22.571342084258088, 'char_order': 6, 'word_order': 0, 'beta': 2}
1.9143	2.0	8630	1.2319	6.1128	{'rouge1': 0.263256301663898, 'rouge2': 0.14256738224583015, 'rougeL': 0.261282034035635, 'rougeLsum': 0.2613517649673947}	{'score': 23.235214776547263, 'char_order': 6, 'word_order': 0, 'beta': 2}
1.8644	3.0	12945	1.2192	6.2145	{'rouge1': 0.2670714744552978, 'rouge2': 0.14606073298261613, 'rougeL': 0.2652594809906982, 'rougeLsum': 0.26489596193447795}	{'score': 23.438449086905997, 'char_order': 6, 'word_order': 0, 'beta': 2}
1.8539	4.0	17260	1.2179	6.2043	{'rouge1': 0.26678061058524805, 'rouge2': 0.14565482302690236, 'rougeL': 0.26489350144733725, 'rougeLsum': 0.26477198178581135}	{'score': 23.464895899326955, 'char_order': 6, 'word_order': 0, 'beta': 2}
1.8525	5.0	21575	1.2179	6.2035	{'rouge1': 0.2667970960136926, 'rouge2': 0.14574925525428614, 'rougeL': 0.26511828595423204, 'rougeLsum': 0.26501665904942706}	{'score': 23.454551827072866, '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