Version2ASAP_FineTuningBERT_AugV8_k5_task1_organization_k5_k5_fold1

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8748
• Qwk: 0.3710
• Mse: 0.8733
• Rmse: 0.9345

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use 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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	1.0	4	9.2544	0.0	9.2518	3.0417
No log	2.0	8	5.9116	0.0010	5.9093	2.4309
No log	3.0	12	3.2453	0.0	3.2435	1.8010
No log	4.0	16	1.9404	0.0810	1.9386	1.3923
No log	5.0	20	1.1837	0.0	1.1823	1.0873
No log	6.0	24	1.4152	0.0	1.4138	1.1890
No log	7.0	28	1.1170	0.0	1.1157	1.0562
No log	8.0	32	1.6021	0.0211	1.6006	1.2651
No log	9.0	36	1.8942	0.0344	1.8927	1.3758
No log	10.0	40	0.9301	0.0618	0.9289	0.9638
No log	11.0	44	1.4030	0.0420	1.4018	1.1840
No log	12.0	48	0.8723	0.1600	0.8712	0.9334
No log	13.0	52	1.0218	0.1003	1.0205	1.0102
No log	14.0	56	0.8248	0.4140	0.8238	0.9076
No log	15.0	60	1.0315	0.0680	1.0302	1.0150
No log	16.0	64	0.7981	0.2511	0.7970	0.8928
No log	17.0	68	0.8125	0.3937	0.8113	0.9007
No log	18.0	72	0.7913	0.4609	0.7901	0.8889
No log	19.0	76	0.7895	0.5681	0.7888	0.8881
No log	20.0	80	0.8706	0.4852	0.8694	0.9324
No log	21.0	84	0.7570	0.4544	0.7559	0.8694
No log	22.0	88	0.8443	0.3958	0.8431	0.9182
No log	23.0	92	0.8437	0.4123	0.8424	0.9178
No log	24.0	96	0.8361	0.4618	0.8346	0.9135
No log	25.0	100	0.7967	0.4858	0.7953	0.8918
No log	26.0	104	0.7781	0.4268	0.7768	0.8814
No log	27.0	108	0.9595	0.3371	0.9580	0.9788
No log	28.0	112	0.8762	0.3972	0.8748	0.9353
No log	29.0	116	0.7565	0.4558	0.7551	0.8690
No log	30.0	120	0.8749	0.4205	0.8733	0.9345
No log	31.0	124	0.9010	0.4104	0.8994	0.9483
No log	32.0	128	0.8304	0.4035	0.8289	0.9104
No log	33.0	132	0.7351	0.4442	0.7338	0.8566
No log	34.0	136	0.9868	0.3065	0.9853	0.9926
No log	35.0	140	1.1532	0.1881	1.1513	1.0730
No log	36.0	144	0.8888	0.3443	0.8872	0.9419
No log	37.0	148	0.8048	0.4117	0.8034	0.8963
No log	38.0	152	0.8155	0.4065	0.8140	0.9022
No log	39.0	156	0.8748	0.3710	0.8733	0.9345

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0