fine-tuned-QAS-Squad_2-with-xlm-roberta-large

This model is a fine-tuned version of xlm-roberta-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8615
• Exact Match: 69.2340
• F1: 82.5542

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Exact Match	F1
1.0967	0.5	464	1.0076	57.8908	71.8971
0.912	1.0	928	0.8118	65.0306	79.0193
0.8071	1.5	1392	0.7587	67.2288	80.3986
0.7414	2.0	1856	0.7322	68.3614	81.3949
0.6548	2.5	2320	0.7685	67.5896	81.3012
0.624	3.0	2784	0.7307	68.5544	82.0875
0.5412	3.5	3248	0.7606	69.2340	82.4384
0.5356	4.0	3712	0.7352	69.5612	82.7509
0.4463	4.5	4176	0.7862	69.2843	82.3298
0.4899	5.0	4640	0.7868	69.5109	82.7397
0.417	5.5	5104	0.8615	69.2340	82.5542

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu117
• Datasets 2.2.0
• Tokenizers 0.13.2