bert-uncased-finetuned-csv-qa

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

• Loss: 0.1688

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	2	5.4411
No log	2.0	4	4.5683
No log	3.0	6	3.9019
No log	4.0	8	3.3646
No log	5.0	10	2.8095
No log	6.0	12	2.3153
No log	7.0	14	1.9456
No log	8.0	16	1.6904
No log	9.0	18	1.4633
No log	10.0	20	1.2169
No log	11.0	22	0.9963
No log	12.0	24	0.8378
No log	13.0	26	0.7068
No log	14.0	28	0.6008
No log	15.0	30	0.5232
No log	16.0	32	0.4659
No log	17.0	34	0.4140
No log	18.0	36	0.3740
No log	19.0	38	0.3459
No log	20.0	40	0.3142
No log	21.0	42	0.2912
No log	22.0	44	0.2766
No log	23.0	46	0.2710
No log	24.0	48	0.2711
No log	25.0	50	0.2730
No log	26.0	52	0.2645
No log	27.0	54	0.2507
No log	28.0	56	0.2432
No log	29.0	58	0.2373
No log	30.0	60	0.2405
No log	31.0	62	0.2397
No log	32.0	64	0.2352
No log	33.0	66	0.2298
No log	34.0	68	0.2255
No log	35.0	70	0.2191
No log	36.0	72	0.2129
No log	37.0	74	0.2068
No log	38.0	76	0.2039
No log	39.0	78	0.2036
No log	40.0	80	0.2004
No log	41.0	82	0.2030
No log	42.0	84	0.2003
No log	43.0	86	0.1958
No log	44.0	88	0.1915
No log	45.0	90	0.1868
No log	46.0	92	0.1812
No log	47.0	94	0.1769
No log	48.0	96	0.1729
No log	49.0	98	0.1698
No log	50.0	100	0.1688

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cpu
• Datasets 3.1.0
• Tokenizers 0.20.3