README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - tatoeba
metrics:
  - bleu
model-index:
  - name: t5-small-finetuned-es-to-pt
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: tatoeba
          type: tatoeba
          args: es-pt
        metrics:
          - name: Bleu
            type: bleu
            value: 15.0473

t5-small-finetuned-es-to-pt

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

• Loss: 1.5557
• Bleu: 15.0473
• Gen Len: 15.8693

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
2.2027	1.0	1907	1.7884	11.6192	15.8829
1.9296	2.0	3814	1.6034	14.201	15.8935
1.8364	3.0	5721	1.5557	15.0473	15.8693

Framework versions

• Transformers 4.17.0
• Pytorch 1.10.0+cu111
• Datasets 1.18.4
• Tokenizers 0.11.6