bert2bert-Large / README.md
Alfahluzi's picture
Update README.md
5447426 verified
metadata
tags:
  - generated_from_trainer
datasets:
  - id_liputan6
model-index:
  - name: >-
      bert2bert-extremecleandata-lr-5e-05-batchsize-2-encmaxlen-2048-decmaxlen-512-abs
    results: []

bert2bert-extremecleandata-lr-5e-05-batchsize-2-encmaxlen-2048-decmaxlen-512-abs

  • Dev set: Extreme clean data
  • Encoder max length (input): 2048
  • Decoder max length (output): 512

This model was trained from scratch on the id_liputan6 dataset. It achieves the following results on the evaluation set:

  • Loss: 9.4148
  • R1 Precision: 0.0188
  • R1 Recall: 0.0105
  • R1 Fmeasure: 0.0133
  • R2 Precision: 0.0
  • R2 Recall: 0.0
  • R2 Fmeasure: 0.0
  • Rl Precision: 0.0188
  • Rl Recall: 0.0106
  • Rl Fmeasure: 0.0133

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: 2
  • eval_batch_size: 2
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss R1 Precision R1 Recall R1 Fmeasure R2 Precision R2 Recall R2 Fmeasure Rl Precision Rl Recall Rl Fmeasure
7.0391 1.0 96942 8.9423 0.0188 0.0105 0.0133 0.0 0.0 0.0 0.0188 0.0106 0.0133
7.0611 2.0 193884 8.9023 0.0188 0.0105 0.0133 0.0 0.0 0.0 0.0188 0.0106 0.0133
7.0266 3.0 290826 9.4047 0.0188 0.0105 0.0133 0.0 0.0 0.0 0.0188 0.0106 0.0133
7.0237 4.0 387768 9.2888 0.0188 0.0105 0.0133 0.0 0.0 0.0 0.0188 0.0106 0.0133
6.9911 5.0 484710 9.4148 0.0188 0.0105 0.0133 0.0 0.0 0.0 0.0188 0.0106 0.0133

Framework versions

  • Transformers 4.39.0
  • Pytorch 2.2.1
  • Datasets 2.18.0
  • Tokenizers 0.15.2