Edit model card

AraElectra-finetuned-fnd

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

  • Loss: 0.6073
  • Macro F1: 0.7629
  • Accuracy: 0.7708
  • Precision: 0.7646
  • Recall: 0.7616

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: 16
  • eval_batch_size: 8
  • seed: 25
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 4

Training results

Training Loss Epoch Step Validation Loss Macro F1 Accuracy Precision Recall
0.5248 1.0 1597 0.4960 0.7416 0.7546 0.7508 0.7377
0.4308 2.0 3194 0.4770 0.7535 0.7666 0.7647 0.7490
0.3386 3.0 4791 0.5201 0.7614 0.7684 0.7617 0.7611
0.2781 4.0 6388 0.6073 0.7629 0.7708 0.7646 0.7616

Framework versions

  • Transformers 4.27.4
  • Pytorch 1.13.0
  • Datasets 2.1.0
  • Tokenizers 0.13.2
Downloads last month
101
Inference API
This model can be loaded on Inference API (serverless).