t5-mt-en-fr

This model is a fine-tuned version of t5-small on the opus_books dataset. It achieves the following results on the evaluation set:

• Loss: 1.4927
• Bleu: 6.3847
• Gen Len: 17.5615

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: 2e-05
• 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: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
1.7324	1.0	6355	1.5504	6.013	17.5801
1.7308	2.0	12710	1.5209	6.2068	17.5694
1.6891	3.0	19065	1.5039	6.2946	17.5627
1.6755	4.0	25420	1.4950	6.3487	17.5619
1.6828	5.0	31775	1.4927	6.3847	17.5615

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3