Edit model card

type-prediction-transformer

This model is a fine-tuned version of airesearch/wangchanberta-base-att-spm-uncased on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0497
  • F1: 0.8651
  • Roc Auc: 0.9260
  • Accuracy: 0.8208

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: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 8

Training results

Training Loss Epoch Step Validation Loss F1 Roc Auc Accuracy
No log 1.0 149 0.0812 0.8070 0.8677 0.7588
No log 2.0 298 0.0591 0.8585 0.9064 0.8141
No log 3.0 447 0.0493 0.8719 0.9144 0.8258
0.0886 4.0 596 0.0506 0.8614 0.9222 0.8090
0.0886 5.0 745 0.0487 0.8683 0.9255 0.8174
0.0886 6.0 894 0.0506 0.8693 0.9291 0.8191
0.0254 7.0 1043 0.0519 0.8619 0.9307 0.8090
0.0254 8.0 1192 0.0497 0.8651 0.9260 0.8208

Framework versions

  • Transformers 4.28.0
  • Pytorch 2.0.0+cu118
  • Tokenizers 0.13.3
Downloads last month
12
Safetensors
Model size
105M params
Tensor type
I64
·
F32
·