README.md

metadata

license: apache-2.0
base_model: t5-small
tags:
  - generated_from_trainer
datasets:
  - opus_books
metrics:
  - bleu
model-index:
  - name: my_awesome_opus_books_model
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: opus_books
          type: opus_books
          config: en-fr
          split: train
          args: en-fr
        metrics:
          - name: Bleu
            type: bleu
            value: 6.1787

my_awesome_opus_books_model

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: 1.5208
• Bleu: 6.1787
• Gen Len: 17.5934

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: 64
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
1.9112	1.0	1589	1.6537	5.2795	17.6478
1.8421	2.0	3178	1.6098	5.6138	17.6216
1.8012	3.0	4767	1.5801	5.7983	17.6147
1.776	4.0	6356	1.5611	5.9587	17.6061
1.7581	5.0	7945	1.5474	6.0336	17.5977
1.7416	6.0	9534	1.5368	6.0882	17.5966
1.7261	7.0	11123	1.5297	6.1366	17.5969
1.7279	8.0	12712	1.5245	6.1442	17.5948
1.7112	9.0	14301	1.5217	6.1715	17.5931
1.7162	10.0	15890	1.5208	6.1787	17.5934

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0