ASAP_FineTuningBERT_AugV5_k1_task1_organization_fold4

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: 0.8594
• Qwk: 0.3031
• Mse: 0.8594
• Rmse: 0.9270

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	2.0	2	9.7069	0.0018	9.7069	3.1156
No log	4.0	4	8.0136	0.0018	8.0136	2.8308
No log	6.0	6	6.4826	0.0018	6.4826	2.5461
No log	8.0	8	5.2953	0.0408	5.2953	2.3011
5.0186	10.0	10	4.1384	0.0118	4.1384	2.0343
5.0186	12.0	12	3.5945	0.0118	3.5945	1.8959
5.0186	14.0	14	2.7705	0.0040	2.7705	1.6645
5.0186	16.0	16	2.2841	0.1614	2.2841	1.5113
5.0186	18.0	18	2.1905	0.1047	2.1905	1.4800
2.1604	20.0	20	1.6308	0.0420	1.6308	1.2770
2.1604	22.0	22	1.5246	0.0420	1.5246	1.2347
2.1604	24.0	24	1.3066	0.0316	1.3066	1.1431
2.1604	26.0	26	1.3203	0.0420	1.3203	1.1490
2.1604	28.0	28	1.4331	0.0420	1.4331	1.1971
1.7805	30.0	30	1.6501	0.0342	1.6501	1.2846
1.7805	32.0	32	1.6684	0.0381	1.6684	1.2917
1.7805	34.0	34	1.4198	0.0445	1.4198	1.1915
1.7805	36.0	36	1.4007	0.0456	1.4007	1.1835
1.7805	38.0	38	1.3889	0.0656	1.3889	1.1785
1.4513	40.0	40	1.3126	0.0697	1.3126	1.1457
1.4513	42.0	42	1.1941	0.0571	1.1941	1.0927
1.4513	44.0	44	1.1476	0.0748	1.1476	1.0713
1.4513	46.0	46	1.0208	0.1117	1.0208	1.0104
1.4513	48.0	48	1.2392	0.1751	1.2392	1.1132
0.8434	50.0	50	1.3578	0.1877	1.3578	1.1652
0.8434	52.0	52	1.0215	0.2745	1.0215	1.0107
0.8434	54.0	54	1.0940	0.2982	1.0940	1.0459
0.8434	56.0	56	0.9441	0.3156	0.9441	0.9717
0.8434	58.0	58	1.0797	0.3110	1.0797	1.0391
0.4526	60.0	60	0.9811	0.3059	0.9811	0.9905
0.4526	62.0	62	0.9059	0.3465	0.9059	0.9518
0.4526	64.0	64	0.9486	0.3205	0.9486	0.9740
0.4526	66.0	66	0.7783	0.3387	0.7783	0.8822
0.4526	68.0	68	0.8113	0.3253	0.8113	0.9007
0.2944	70.0	70	1.0263	0.3179	1.0263	1.0130
0.2944	72.0	72	0.8863	0.3409	0.8863	0.9414
0.2944	74.0	74	0.7964	0.3267	0.7964	0.8924
0.2944	76.0	76	0.8521	0.3294	0.8521	0.9231
0.2944	78.0	78	0.9574	0.3250	0.9574	0.9784
0.211	80.0	80	0.8926	0.3277	0.8926	0.9448
0.211	82.0	82	0.8061	0.3094	0.8061	0.8979
0.211	84.0	84	0.8071	0.3123	0.8071	0.8984
0.211	86.0	86	0.8848	0.3151	0.8848	0.9407
0.211	88.0	88	0.9816	0.3239	0.9816	0.9908
0.1858	90.0	90	0.9544	0.3360	0.9544	0.9769
0.1858	92.0	92	0.8603	0.3031	0.8603	0.9275
0.1858	94.0	94	0.8166	0.3222	0.8166	0.9037
0.1858	96.0	96	0.8203	0.3250	0.8203	0.9057
0.1858	98.0	98	0.8426	0.3140	0.8426	0.9179
0.1627	100.0	100	0.8594	0.3031	0.8594	0.9270

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1