README.md

metadata

license: apache-2.0
base_model: t5-small
tags:
  - generated_from_trainer
datasets:
  - opus_books
metrics:
  - bleu
model-index:
  - name: opus_books_es_pt
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: opus_books
          type: opus_books
          config: es-pt
          split: train
          args: es-pt
        metrics:
          - name: Bleu
            type: bleu
            value: 0.4173

opus_books_es_pt

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: 3.4106
• Bleu: 0.4173
• Gen Len: 18.0263

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
No log	1.0	67	3.7925	0.0732	17.9511
No log	2.0	134	3.5838	0.1097	17.9774
No log	3.0	201	3.4790	0.29	17.9586
No log	4.0	268	3.4281	0.3982	18.0
No log	5.0	335	3.4106	0.4173	18.0263

Framework versions

• Transformers 4.36.1
• Pytorch 2.1.0+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0