augmented_t5_pictos / README.md
santyzenith's picture
End of training
7d13e88
|
raw
history blame
2.52 kB
metadata
license: mit
base_model: flax-community/spanish-t5-small
tags:
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: augmented_t5_pictos
    results: []

augmented_t5_pictos

This model is a fine-tuned version of flax-community/spanish-t5-small on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 1.4052
  • Rouge1: 42.0934
  • Rouge2: 28.3804
  • Rougel: 41.2489
  • Rougelsum: 41.3148
  • Gen Len: 8.4979

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
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
2.3625 1.0 527 2.0249 34.7282 19.5441 33.9362 33.9053 7.6303
1.9453 2.0 1054 1.7992 37.5587 22.5669 36.6904 36.7767 8.2671
1.7429 3.0 1581 1.6655 39.7178 24.1301 38.8293 38.8615 8.3462
1.6306 4.0 2108 1.5791 40.8591 26.0948 40.088 40.0812 7.9573
1.461 5.0 2635 1.5204 40.6998 25.5825 39.8972 39.9642 8.3568
1.4178 6.0 3162 1.4702 41.0675 26.4585 40.1879 40.2333 8.4679
1.3394 7.0 3689 1.4452 41.9649 27.1925 41.0945 41.1233 8.2329
1.2844 8.0 4216 1.4210 41.9633 27.7102 40.9804 41.0706 8.5406
1.2151 9.0 4743 1.4072 41.9336 28.0917 41.0597 41.1328 8.4722
1.215 10.0 5270 1.4052 42.0934 28.3804 41.2489 41.3148 8.4979

Framework versions

  • Transformers 4.34.1
  • Pytorch 2.1.0+cu118
  • Datasets 2.14.6
  • Tokenizers 0.14.1