roberta-vmw-mrqa-s

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

• Loss: 1.2333

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: 20
• eval_batch_size: 20
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.03
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss
1.149	0.0357	50	1.2558
1.1432	0.0715	100	1.2662
1.1559	0.1072	150	1.2555
1.2238	0.1430	200	1.2339
1.1742	0.1787	250	1.2422
1.169	0.2144	300	1.2393
1.1487	0.2502	350	1.2296
1.163	0.2859	400	1.2310
1.1779	0.3217	450	1.2311
1.1303	0.3574	500	1.2333

Framework versions

• Transformers 4.42.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1