Hugging Face's logo

vania2911
/
5661

Text2Text Generation
Transformers
TensorBoard
Safetensors
marian
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
5661 / README.md
vania2911's picture
vania2911
End of training
77a13e9 verified
preview code
|
raw
history blame contribute delete
8.63 kB
metadata
library_name: transformers
license: apache-2.0
base_model: Helsinki-NLP/opus-mt-es-es
tags:
  - generated_from_trainer
model-index:
  - name: '5661'
    results: []

5661

This model is a fine-tuned version of Helsinki-NLP/opus-mt-es-es on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5476
  • Model Preparation Time: 0.0034
  • Bleu Msl: 0.0
  • Bleu 1 Msl: 0.6746
  • Bleu 2 Msl: 0.0151
  • Bleu 3 Msl: 0.0045
  • Bleu 4 Msl: 0.0023
  • Ter Msl: 100
  • Bleu Asl: 0
  • Bleu 1 Asl: 0
  • Bleu 2 Asl: 0
  • Bleu 3 Asl: 0
  • Bleu 4 Asl: 0
  • Ter Asl: 100

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: 0.0001
  • train_batch_size: 32
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 30
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Model Preparation Time Bleu Msl Bleu 1 Msl Bleu 2 Msl Bleu 3 Msl Bleu 4 Msl Ter Msl Bleu Asl Bleu 1 Asl Bleu 2 Asl Bleu 3 Asl Bleu 4 Asl Ter Asl
No log 1.0 142 0.3764 0.0034 35.3553 0.5627 0.0104 0.0029 0.0014 100 100.0000 0.4152 0.0083 0.0024 0.0012 100
No log 2.0 284 0.3592 0.0034 37.9918 0.5665 0.0104 0.0029 0.0014 100 53.7285 0.3970 0.0081 0.0024 0.0012 100
No log 3.0 426 0.3690 0.0034 50.0000 0.5627 0.0104 0.0029 0.0014 100 32.1729 0.3427 0.0075 0.0022 0.0011 100
0.019 4.0 568 0.3660 0.0034 50.0000 0.5875 0.0106 0.0029 0.0014 100 37.9918 0.4135 0.0083 0.0024 0.0012 100
0.019 5.0 710 0.3824 0.0034 50.0000 0.6350 0.0110 0.0030 0.0014 100 100.0000 0.4465 0.0086 0.0024 0.0012 100
0.019 6.0 852 0.3574 0.0034 50.0000 0.5019 0.0098 0.0028 0.0014 100 42.7287 0.4217 0.0083 0.0024 0.0012 100
0.019 7.0 994 0.3810 0.0034 0.0 0.5361 0.0101 0.0029 0.0014 100 100.0000 0.4300 0.0084 0.0024 0.0012 100
0.0188 8.0 1136 0.3659 0.0034 50.0000 0.5 0.0098 0.0028 0.0014 100 100.0000 0.3904 0.0080 0.0023 0.0012 100
0.0188 9.0 1278 0.3696 0.0034 50.0000 0.6236 0.0109 0.0030 0.0014 100 100.0000 0.4778 0.0089 0.0025 0.0012 100
0.0188 10.0 1420 0.3749 0.0034 50.0000 0.5970 0.0107 0.0030 0.0014 100 100.0000 0.4498 0.0086 0.0025 0.0012 100
0.0107 11.0 1562 0.3746 0.0034 50.0000 0.5228 0.0100 0.0028 0.0014 100 70.7107 0.4349 0.0085 0.0024 0.0012 100
0.0107 12.0 1704 0.3635 0.0034 50.0000 0.5627 0.0104 0.0029 0.0014 100 100.0000 0.4728 0.0088 0.0025 0.0012 100
0.0107 13.0 1846 0.3629 0.0034 50.0000 0.5856 0.0106 0.0029 0.0014 100 100.0000 0.4481 0.0086 0.0024 0.0012 100
0.0107 14.0 1988 0.3697 0.0034 50.0000 0.5760 0.0105 0.0029 0.0014 100 100.0000 0.4778 0.0089 0.0025 0.0012 100
0.0096 15.0 2130 0.3630 0.0034 50.0000 0.5570 0.0103 0.0029 0.0014 100 59.4604 0.4514 0.0086 0.0025 0.0012 100
0.0096 16.0 2272 0.3608 0.0034 50.0000 0.5304 0.0101 0.0028 0.0014 100 100.0000 0.4481 0.0086 0.0024 0.0012 100
0.0096 17.0 2414 0.3612 0.0034 50.0000 0.5646 0.0104 0.0029 0.0014 100 100.0000 0.4415 0.0085 0.0024 0.0012 100
0.0057 18.0 2556 0.3571 0.0034 50.0000 0.5646 0.0104 0.0029 0.0014 100 100.0000 0.4415 0.0085 0.0024 0.0012 100
0.0057 19.0 2698 0.3655 0.0034 50.0000 0.5817 0.0105 0.0029 0.0014 100 100.0000 0.4514 0.0086 0.0025 0.0012 100
0.0057 20.0 2840 0.3682 0.0034 50.0000 0.5760 0.0105 0.0029 0.0014 100 100.0000 0.4580 0.0087 0.0025 0.0012 100
0.0057 21.0 2982 0.3683 0.0034 50.0000 0.5722 0.0104 0.0029 0.0014 100 100.0000 0.4646 0.0088 0.0025 0.0012 100
0.0046 22.0 3124 0.3745 0.0034 50.0000 0.5798 0.0105 0.0029 0.0014 100 100.0000 0.4695 0.0088 0.0025 0.0012 100
0.0046 23.0 3266 0.3680 0.0034 50.0000 0.6198 0.0109 0.0030 0.0014 100 100.0000 0.4811 0.0089 0.0025 0.0012 100
0.0046 24.0 3408 0.3702 0.0034 50.0000 0.5951 0.0106 0.0030 0.0014 100 100.0000 0.4811 0.0089 0.0025 0.0012 100
0.0039 25.0 3550 0.3701 0.0034 50.0000 0.5951 0.0106 0.0030 0.0014 100 100.0000 0.4959 0.0090 0.0025 0.0012 100
0.0039 26.0 3692 0.3689 0.0034 50.0000 0.6065 0.0107 0.0030 0.0014 100 100.0000 0.5041 0.0091 0.0025 0.0012 100
0.0039 27.0 3834 0.3681 0.0034 50.0000 0.6084 0.0108 0.0030 0.0014 100 100.0000 0.5008 0.0091 0.0025 0.0012 100
0.0039 28.0 3976 0.3684 0.0034 50.0000 0.6217 0.0109 0.0030 0.0014 100 100.0000 0.5025 0.0091 0.0025 0.0012 100
0.0033 29.0 4118 0.3702 0.0034 50.0000 0.6122 0.0108 0.0030 0.0014 100 100.0000 0.5008 0.0091 0.0025 0.0012 100
0.0033 30.0 4260 0.3700 0.0034 50.0000 0.6160 0.0108 0.0030 0.0014 100 100.0000 0.5008 0.0091 0.0025 0.0012 100

Framework versions

  • Transformers 4.49.0
  • Pytorch 2.5.1+cu124
  • Datasets 3.3.2
  • Tokenizers 0.21.0