distilroberta-base-finetuned-squad2-lwt

This model is a fine-tuned version of distilroberta-base on the squad_v2 dataset. It achieves the following results on the evaluation set:

• Loss: 1.1356

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
1.1702	1.0	4120	1.1220
0.9787	2.0	8240	1.0500
0.8153	3.0	12360	1.1356

Framework versions

• Transformers 4.17.0
• Pytorch 1.10.0+cu111
• Datasets 2.0.0
• Tokenizers 0.11.6

{'HasAns_exact': 71.39001349527665, 'HasAns_f1': 77.71740687727831, 'HasAns_total': 5928, 'NoAns_exact': 68.59545836837678, 'NoAns_f1': 68.59545836837678, 'NoAns_total': 5945, 'best_exact': 69.9991577528847, 'best_exact_thresh': 0.0, 'best_f1': 73.1583245993857, 'best_f1_thresh': 0.0, 'exact': 69.99073528173166, 'f1': 73.1499021282327, 'total': 11873}