bert-10

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: 9.5797

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: 8
• eval_batch_size: 8
• 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
10.8556	0.05	5	12.3235
10.8413	0.09	10	12.2591
11.0649	0.14	15	12.1778
11.6408	0.18	20	12.0989
11.3732	0.23	25	12.0213
10.5122	0.28	30	11.9458
10.6594	0.32	35	11.8691
10.745	0.37	40	11.7928
10.8256	0.41	45	11.7163
10.1627	0.46	50	11.6430
10.9907	0.5	55	11.5703
10.1394	0.55	60	11.4997
9.6059	0.6	65	11.4287
9.4972	0.64	70	11.3621
10.2252	0.69	75	11.2949
10.4887	0.73	80	11.2288
9.9616	0.78	85	11.1638
9.5775	0.83	90	11.1003
9.5971	0.87	95	11.0381
9.5745	0.92	100	10.9773
9.3218	0.96	105	10.9178
9.4906	1.01	110	10.8597
9.1168	1.06	115	10.8030
9.8009	1.1	120	10.7465
9.3632	1.15	125	10.6915
8.9858	1.19	130	10.6399
9.2904	1.24	135	10.5874
9.5344	1.28	140	10.5370
9.0034	1.33	145	10.4871
9.3024	1.38	150	10.4384
8.7905	1.42	155	10.3920
8.9329	1.47	160	10.3465
8.9834	1.51	165	10.3027
8.7307	1.56	170	10.2607
8.6729	1.61	175	10.2200
9.1849	1.65	180	10.1794
9.1618	1.7	185	10.1400
8.9048	1.74	190	10.1023
8.9427	1.79	195	10.0655
9.1052	1.83	200	10.0294
9.1123	1.88	205	9.9938
9.0476	1.93	210	9.9604
8.5532	1.97	215	9.9285
8.7871	2.02	220	9.8977
8.5984	2.06	225	9.8690
8.7009	2.11	230	9.8414
8.9376	2.16	235	9.8146
8.3535	2.2	240	9.7906
8.5805	2.25	245	9.7675
8.4641	2.29	250	9.7463
8.3975	2.34	255	9.7263
8.7698	2.39	260	9.7070
8.3541	2.43	265	9.6901
8.5443	2.48	270	9.6743
8.1539	2.52	275	9.6595
7.9856	2.57	280	9.6459
8.2532	2.61	285	9.6333
8.2116	2.66	290	9.6221
8.9557	2.71	295	9.6119
8.0754	2.75	300	9.6032
7.9534	2.8	305	9.5956
8.5578	2.84	310	9.5899
8.6403	2.89	315	9.5848
8.1103	2.94	320	9.5817
8.3785	2.98	325	9.5797

Framework versions

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