vgaraujov's picture
Update README.md
8ba1daa
metadata
license: apache-2.0
language:
  - en
  - es
base_model: vgaraujov/t5-base-spanish
tags:
  - generated_from_trainer
datasets:
  - vgaraujov/wmt13
metrics:
  - bleu
model-index:
  - name: t5-base-translation-en-es
    results:
      - task:
          name: Translation
          type: translation
        dataset:
          name: vgaraujov/wmt13 es-en
          type: vgaraujov/wmt13
          config: es-en
          split: validation
          args: es-en
        metrics:
          - name: Bleu
            type: bleu
            value: 30.6296
widget:
  - text: Hey, I am T5S for translation.

T5S (base-sized model) for en-es translation

This model is a fine-tuned version of T5S on a small portion of WMT13 es-en dataset. It achieves the following results on the evaluation set:

  • Loss: 1.7643
  • Bleu: 30.6296
  • Gen Len: 29.2701

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.005
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 12
  • total_train_batch_size: 384
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 40000
  • training_steps: 10000

Framework versions

  • Transformers 4.33.0.dev0
  • Pytorch 2.0.1+cu117
  • Datasets 2.14.4
  • Tokenizers 0.13.3