PaulineSanchez's picture
Upload 3 files
bda467e
|
raw
history blame
2 kB
metadata
language:
  - fr
  - en
license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - PaulineSanchez/recipes_translation_400
metrics:
  - bleu
model-index:
  - name: train_hf_new_fren_batch_12_epoch_4
    results:
      - task:
          name: Translation
          type: translation
        dataset:
          name: PaulineSanchez/recipes_translation_400
          type: PaulineSanchez/recipes_translation_400
          split: validation
        metrics:
          - name: Bleu
            type: bleu
            value: 59.7829

train_hf_new_fren_batch_12_epoch_4

This model is a fine-tuned version of Helsinki-NLP/opus-mt-fr-en on the PaulineSanchez/recipes_translation_400 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7762
  • Bleu: 59.7829
  • Gen Len: 29.475

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: 5e-05
  • train_batch_size: 12
  • eval_batch_size: 12
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 4.0

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
No log 1.0 27 0.8249 57.2256 29.225
No log 2.0 54 0.7964 58.125 28.95
No log 3.0 81 0.7786 59.3353 29.35
No log 4.0 108 0.7762 59.7829 29.475

Framework versions

  • Transformers 4.31.0.dev0
  • Pytorch 2.0.1+cu117
  • Datasets 2.12.0
  • Tokenizers 0.13.3