Model4_arabertv2_base_T1_WS_A100_2nd

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02-twitter on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1510
• F1 Micro: 0.8382
• F1 Macro: 0.7689
• Roc Auc: 0.8969
• Accuracy: 0.8024

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Micro	F1 Macro	Roc Auc	Accuracy
0.1687	1.0	507	0.1510	0.8382	0.7689	0.8969	0.8024
0.0892	2.0	1014	0.1634	0.8337	0.7637	0.8925	0.7982
0.0539	3.0	1521	0.1884	0.8318	0.7630	0.8968	0.7996
0.0331	4.0	2028	0.2067	0.8419	0.7700	0.9041	0.8108

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.20.3