Broomva's picture
End of training
618c9f7
metadata
license: apache-2.0
base_model: google/mt5-base
tags:
  - generated_from_trainer
metrics:
  - bleu
model-index:
  - name: mt5-base-translation-spa-guc
    results: []

mt5-base-translation-spa-guc

This model is a fine-tuned version of google/mt5-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.4465
  • Bleu: 3.088
  • Gen Len: 16.5468

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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 10
  • num_epochs: 15

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
2.3515 1.0 7668 2.3255 1.2092 16.8878
2.3229 2.0 15336 2.0343 1.3948 16.6967
1.828 3.0 23004 1.8707 1.7948 16.729
1.7692 4.0 30672 1.7654 1.9816 16.5538
3.1073 5.0 38340 1.6878 2.2335 16.6662
1.8726 6.0 46008 1.6280 2.4231 16.566
1.8611 7.0 53676 1.5870 2.5839 16.5724
1.4777 8.0 61344 1.5476 2.7415 16.5651
1.5331 9.0 69012 1.5181 2.8105 16.5862
1.6348 10.0 76680 1.4972 2.9386 16.582
1.2195 11.0 84348 1.4739 2.9339 16.578
1.8651 12.0 92016 1.4647 3.0058 16.5431
1.3484 13.0 99684 1.4537 3.079 16.5513
1.5675 14.0 107352 1.4484 3.0925 16.5496
1.6518 15.0 115020 1.4465 3.088 16.5468

Framework versions

  • Transformers 4.35.2
  • Pytorch 1.13.1+cu117
  • Datasets 2.15.0
  • Tokenizers 0.15.0