Finetune_XLM_R_large_QA

This model is a fine-tuned version of FacebookAI/xlm-roberta-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.8605
• Exact: 42.4751
• F1: 46.2340
• Total: 3814
• Hasans Exact: 17.3012
• Hasans F1: 22.7051
• Hasans Total: 2653
• Noans Exact: 100.0
• Noans F1: 100.0
• Noans Total: 1161
• Best Exact: 42.4751
• Best Exact Thresh: 0.0
• Best F1: 46.2340
• Best F1 Thresh: 0.0

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: 12
• eval_batch_size: 12
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 24
• 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Exact	F1	Total	Hasans Exact	Hasans F1	Hasans Total	Noans Exact	Noans F1	Noans Total	Best Exact	Best Exact Thresh	Best F1	Best F1 Thresh
1.1876	1.0	1262	1.1780	41.8458	45.0678	3814	16.3965	21.0285	2653	100.0	100.0	1161	41.8458	0.0	45.0678	0.0
0.944	2.0	2524	1.1949	42.4227	46.2810	3814	17.2258	22.7727	2653	100.0	100.0	1161	42.4227	0.0	46.2810	0.0
0.6202	3.0	3786	1.2508	43.0781	46.9167	3814	18.1681	23.6865	2653	100.0	100.0	1161	43.0781	0.0	46.9167	0.0
0.4526	4.0	5048	1.3825	42.2391	46.0137	3814	16.9619	22.3884	2653	100.0	100.0	1161	42.2391	0.0	46.0137	0.0
0.2965	5.0	6310	1.6191	42.1342	46.1030	3814	16.8112	22.5167	2653	100.0	100.0	1161	42.1342	0.0	46.1030	0.0
0.2125	6.0	7572	1.8605	42.4751	46.2340	3814	17.3012	22.7051	2653	100.0	100.0	1161	42.4751	0.0	46.2340	0.0

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0