nlparabic's picture
End of training
9920a15 verified
metadata
base_model: aubmindlab/aragpt2-base
tags:
  - generated_from_trainer
metrics:
  - bleu
  - rouge
model-index:
  - name: res_nw_yem_aragpt2-base
    results: []

res_nw_yem_aragpt2-base

This model is a fine-tuned version of aubmindlab/aragpt2-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0534
  • Bleu: 0.0428
  • Rouge1: 0.3139
  • Rouge2: 0.1104
  • Rougel: 0.3097

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: 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: 500
  • num_epochs: 20.0

Training results

Training Loss Epoch Step Validation Loss Bleu Rouge1 Rouge2 Rougel
5.8224 1.0 153 0.1129 0.0039 0.0712 0.0029 0.0695
0.1108 2.0 306 0.0691 0.0 0.0951 0.0078 0.0932
0.0775 3.0 459 0.0628 0.0067 0.1291 0.0157 0.1286
0.0678 4.0 612 0.0592 0.0086 0.1524 0.0273 0.1492
0.0603 5.0 765 0.0566 0.0162 0.1919 0.0413 0.1883
0.0547 6.0 918 0.0546 0.0187 0.2239 0.0599 0.2218
0.0498 7.0 1071 0.0540 0.0295 0.2684 0.0733 0.2638
0.0456 8.0 1224 0.0536 0.0292 0.2884 0.0818 0.2841
0.0419 9.0 1377 0.0534 0.0428 0.3139 0.1104 0.3097
0.0385 10.0 1530 0.0534 0.0461 0.3255 0.1118 0.3185
0.0354 11.0 1683 0.0540 0.0473 0.3358 0.1219 0.3288
0.0331 12.0 1836 0.0540 0.0476 0.3483 0.1312 0.3442
0.0308 13.0 1989 0.0552 0.0590 0.3599 0.1439 0.3539
0.0291 14.0 2142 0.0556 0.0625 0.3737 0.1489 0.3670

Framework versions

  • Transformers 4.45.0.dev0
  • Pytorch 2.3.1+cu121
  • Datasets 2.19.2
  • Tokenizers 0.19.1