Finetune_XLM_R_base_QA_NEW

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

• Loss: 1.5708
• Exact: 65.6004
• F1: 74.4082
• Total: 3814
• Hasans Exact: 65.6004
• Hasans F1: 74.4082
• Hasans Total: 3814
• Best Exact: 65.6004
• Best Exact Thresh: 0.0
• Best F1: 74.4082
• 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: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• optimizer: Use OptimizerNames.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	Best Exact	Best Exact Thresh	Best F1	Best F1 Thresh
1.9984	1.0	214	1.9037	59.5700	70.4638	3814	59.5700	70.4638	3814	59.5700	0.0	70.4638	0.0
1.5927	2.0	428	1.6175	63.1883	72.4873	3814	63.1883	72.4873	3814	63.1883	0.0	72.4873	0.0
1.4047	3.0	642	1.5775	66.3083	76.8255	3814	66.3083	76.8255	3814	66.3083	0.0	76.8255	0.0
1.2589	4.0	856	1.5762	68.9827	79.8908	3814	68.9827	79.8908	3814	68.9827	0.0	79.8908	0.0
1.1412	5.0	1070	1.5405	68.2223	78.0453	3814	68.2223	78.0453	3814	68.2223	0.0	78.0453	0.0
1.0846	6.0	1284	1.5708	65.6004	74.4082	3814	65.6004	74.4082	3814	65.6004	0.0	74.4082	0.0

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1