aomar85/improved_stance_detection_v2-fold1

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

improvedstancedetection_v2-fold1

This model is a fine-tuned version of aubmindlab/bert-large-arabertv02-twitter on an unknown dataset. It achieves the following results on the evaluation set:

• —Loss: 0.4010
• —Accuracy: 0.5902
• —Macro F1: 0.5938
• —Weighted F1: 0.5921
• —F1 Pro: 0.5891
• —F1 Against: 0.5590
• —F1 Neutral: 0.6333

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: 3e-05
• —trainbatchsize: 8
• —evalbatchsize: 16
• —seed: 42
• —gradientaccumulationsteps: 4
• —totaltrainbatch_size: 32
• —optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• —lrschedulertype: cosine
• —lrschedulerwarmup_steps: 0.1
• —num_epochs: 15
• —mixedprecisiontraining: Native AMP
• —labelsmoothingfactor: 0.1

Training results

Framework versions

• —Transformers 5.0.0
• —Pytorch 2.9.0+cu128
• —Datasets 4.0.0
• —Tokenizers 0.22.2