Version3ASAP_FineTuningBERT_AugV14_k3_task1_organization_k3_k3_fold3

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: 0.7886
• Qwk: 0.4971
• Mse: 0.7896
• Rmse: 0.8886

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	3	9.9054	0.0	9.9037	3.1470
No log	2.0	6	8.2150	0.0	8.2135	2.8659
No log	3.0	9	5.7233	0.0809	5.7221	2.3921
No log	4.0	12	4.3518	0.0097	4.3508	2.0859
No log	5.0	15	2.7424	0.0149	2.7416	1.6558
No log	6.0	18	1.6750	0.1167	1.6744	1.2940
No log	7.0	21	1.1938	0.0463	1.1933	1.0924
No log	8.0	24	1.0987	0.0463	1.0982	1.0479
No log	9.0	27	1.3326	0.0478	1.3321	1.1541
No log	10.0	30	0.8737	0.2420	0.8734	0.9346
No log	11.0	33	1.0309	0.1044	1.0305	1.0151
No log	12.0	36	1.1689	0.1252	1.1689	1.0811
No log	13.0	39	0.7316	0.3815	0.7321	0.8556
No log	14.0	42	0.8308	0.3731	0.8313	0.9118
No log	15.0	45	0.9309	0.2860	0.9317	0.9652
No log	16.0	48	0.6810	0.3773	0.6821	0.8259
No log	17.0	51	0.8933	0.3018	0.8943	0.9457
No log	18.0	54	0.8574	0.3676	0.8585	0.9265
No log	19.0	57	0.8440	0.3902	0.8453	0.9194
No log	20.0	60	0.7847	0.4793	0.7858	0.8865
No log	21.0	63	1.2122	0.3039	1.2129	1.1013
No log	22.0	66	0.7215	0.5104	0.7224	0.8500
No log	23.0	69	1.1511	0.3292	1.1519	1.0733
No log	24.0	72	0.8207	0.4954	0.8219	0.9066
No log	25.0	75	1.1714	0.3384	1.1724	1.0828
No log	26.0	78	0.8061	0.4835	0.8072	0.8984
No log	27.0	81	1.1129	0.3486	1.1137	1.0553
No log	28.0	84	0.7462	0.5208	0.7472	0.8644
No log	29.0	87	1.0367	0.3835	1.0377	1.0187
No log	30.0	90	0.8083	0.4870	0.8095	0.8997
No log	31.0	93	0.8482	0.4438	0.8494	0.9216
No log	32.0	96	1.0641	0.3798	1.0652	1.0321
No log	33.0	99	0.7656	0.5093	0.7668	0.8756
No log	34.0	102	0.9406	0.4170	0.9416	0.9704
No log	35.0	105	1.0590	0.3793	1.0599	1.0295
No log	36.0	108	0.7559	0.4865	0.7569	0.8700
No log	37.0	111	0.7851	0.4828	0.7862	0.8867
No log	38.0	114	0.7962	0.5062	0.7974	0.8930
No log	39.0	117	0.8686	0.4624	0.8697	0.9326
No log	40.0	120	0.8761	0.4654	0.8772	0.9366
No log	41.0	123	0.9554	0.4481	0.9565	0.9780
No log	42.0	126	0.7880	0.4941	0.7891	0.8883
No log	43.0	129	0.8765	0.4636	0.8777	0.9368
No log	44.0	132	0.8437	0.5429	0.8450	0.9192
No log	45.0	135	0.8479	0.5113	0.8492	0.9215
No log	46.0	138	0.8249	0.5365	0.8261	0.9089
No log	47.0	141	0.8322	0.4554	0.8333	0.9129
No log	48.0	144	0.8692	0.4673	0.8703	0.9329
No log	49.0	147	0.7978	0.4971	0.7990	0.8938
No log	50.0	150	0.8549	0.4826	0.8561	0.9252
No log	51.0	153	0.8124	0.5047	0.8135	0.9019
No log	52.0	156	0.8512	0.4734	0.8523	0.9232
No log	53.0	159	0.7983	0.5327	0.7994	0.8941
No log	54.0	162	0.9116	0.4581	0.9127	0.9554
No log	55.0	165	0.7736	0.5308	0.7747	0.8802
No log	56.0	168	0.7894	0.5153	0.7905	0.8891
No log	57.0	171	0.9737	0.4324	0.9747	0.9873
No log	58.0	174	0.7920	0.5193	0.7931	0.8906
No log	59.0	177	0.8386	0.4964	0.8397	0.9164
No log	60.0	180	0.8410	0.5034	0.8421	0.9177
No log	61.0	183	0.8421	0.4857	0.8432	0.9183
No log	62.0	186	0.8136	0.4838	0.8147	0.9026
No log	63.0	189	0.8495	0.4652	0.8505	0.9222
No log	64.0	192	0.7886	0.4971	0.7896	0.8886

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0