bert-13

This model is a fine-tuned version of deepset/bert-base-cased-squad2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 10.7394

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-07
• train_batch_size: 16
• eval_batch_size: 16
• 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
11.3227	0.09	5	12.3300
11.424	0.18	10	12.2357
10.9635	0.27	15	12.1466
11.0405	0.36	20	12.0590
10.9641	0.45	25	11.9722
10.9497	0.55	30	11.8886
10.1742	0.64	35	11.8071
10.8066	0.73	40	11.7293
10.2595	0.82	45	11.6547
10.1445	0.91	50	11.5820
9.8379	1.0	55	11.5138
10.0327	1.09	60	11.4487
10.1503	1.18	65	11.3867
10.3143	1.27	70	11.3276
9.7859	1.36	75	11.2704
10.2077	1.45	80	11.2162
9.9116	1.55	85	11.1643
9.8015	1.64	90	11.1145
9.5877	1.73	95	11.0686
9.9183	1.82	100	11.0239
9.4468	1.91	105	10.9848
9.7489	2.0	110	10.9489
9.5956	2.09	115	10.9143
9.4762	2.18	120	10.8835
9.5775	2.27	125	10.8555
9.6383	2.36	130	10.8300
9.2386	2.45	135	10.8071
9.6892	2.55	140	10.7877
9.2103	2.64	145	10.7721
9.774	2.73	150	10.7594
9.5202	2.82	155	10.7501
9.4542	2.91	160	10.7430
9.6628	3.0	165	10.7394

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1