t5-small-finetuned-de-en-lr2e-4

This model is a fine-tuned version of t5-small on the wmt14 dataset. It achieves the following results on the evaluation set:

  • Loss: 2.0115
  • Bleu: 9.12
  • Gen Len: 17.4026

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: linear
  • num_epochs: 5
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
No log 1.0 188 2.0701 8.1225 17.4542
No log 2.0 376 2.0316 8.5741 17.4229
2.2224 3.0 564 2.0229 8.9227 17.3703
2.2224 4.0 752 2.0105 9.0764 17.4053
2.2224 5.0 940 2.0115 9.12 17.4026

Framework versions

  • Transformers 4.12.5
  • Pytorch 1.10.0+cu111
  • Datasets 1.16.1
  • Tokenizers 0.10.3
Downloads last month
3
Hosted inference API
This model can be loaded on the Inference API on-demand.

Dataset used to train rossanez/t5-small-finetuned-de-en-lr2e-4

Evaluation results