google-mt5-base-lora-finetuned-persian-news

• Loss: 0.9099
• Rouge1 Precision: 59.4386
• Rouge1 Recall: 60.3359
• Rouge1 F1: 58.6755
• Rouge2 Precision: 42.2757
• Rouge2 Recall: 43.1573
• Rouge2 F1: 41.7994
• Rougel Precision: 51.4347
• Rougel Recall: 52.3226
• Rougel F1: 50.8172

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

• learning_rate: 0.0005
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAFACTOR and the args are: No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 0.1
• num_epochs: 4

Training results

Framework versions

• PEFT 0.20.0
• Transformers 5.14.1
• Pytorch 2.11.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2