Version_Test_ASAP_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: 0.7215
• Qwk: 0.5432
• Mse: 0.7216
• Rmse: 0.8495

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	10.3849	0.0	10.3828	3.2222
No log	2.0	6	8.4120	0.0	8.4104	2.9001
No log	3.0	9	6.2522	0.0231	6.2509	2.5002
No log	4.0	12	4.4285	0.0	4.4274	2.1041
No log	5.0	15	2.9123	0.0	2.9114	1.7063
No log	6.0	18	1.6437	0.0329	1.6431	1.2818
No log	7.0	21	1.1651	0.0202	1.1646	1.0792
No log	8.0	24	0.8840	0.2992	0.8836	0.9400
No log	9.0	27	0.7901	0.2490	0.7898	0.8887
No log	10.0	30	0.7220	0.3016	0.7219	0.8497
No log	11.0	33	0.6114	0.3964	0.6114	0.7820
No log	12.0	36	0.6487	0.5302	0.6489	0.8055
No log	13.0	39	1.4362	0.3321	1.4366	1.1986
No log	14.0	42	0.9632	0.4906	0.9637	0.9817
No log	15.0	45	0.9946	0.4683	0.9950	0.9975
No log	16.0	48	0.7783	0.4980	0.7786	0.8824
No log	17.0	51	0.8812	0.4835	0.8816	0.9389
No log	18.0	54	0.7283	0.5788	0.7284	0.8535
No log	19.0	57	0.9000	0.4836	0.9003	0.9488
No log	20.0	60	0.8130	0.5219	0.8130	0.9017
No log	21.0	63	1.2287	0.4122	1.2289	1.1086
No log	22.0	66	0.8488	0.5268	0.8488	0.9213
No log	23.0	69	0.9886	0.4806	0.9888	0.9944
No log	24.0	72	0.8863	0.4993	0.8866	0.9416
No log	25.0	75	0.7719	0.5450	0.7721	0.8787
No log	26.0	78	0.7664	0.5242	0.7667	0.8756
No log	27.0	81	0.7632	0.5400	0.7635	0.8738
No log	28.0	84	0.9071	0.4925	0.9074	0.9526
No log	29.0	87	0.7829	0.5468	0.7831	0.8850
No log	30.0	90	0.8345	0.5067	0.8349	0.9137
No log	31.0	93	0.7326	0.5600	0.7329	0.8561
No log	32.0	96	0.8275	0.4940	0.8278	0.9098
No log	33.0	99	0.8441	0.5606	0.8440	0.9187
No log	34.0	102	0.8190	0.5705	0.8190	0.9050
No log	35.0	105	0.9452	0.4692	0.9456	0.9724
No log	36.0	108	0.7512	0.5452	0.7515	0.8669
No log	37.0	111	0.7433	0.5836	0.7432	0.8621
No log	38.0	114	0.7195	0.5401	0.7197	0.8484
No log	39.0	117	0.6972	0.5652	0.6972	0.8350
No log	40.0	120	0.7524	0.5369	0.7527	0.8676
No log	41.0	123	0.7182	0.5347	0.7184	0.8476
No log	42.0	126	0.7536	0.5268	0.7538	0.8682
No log	43.0	129	0.7240	0.5503	0.7242	0.8510
No log	44.0	132	0.7028	0.5624	0.7029	0.8384
No log	45.0	135	0.7048	0.5590	0.7050	0.8397
No log	46.0	138	0.7521	0.5384	0.7524	0.8674
No log	47.0	141	0.6997	0.5470	0.7000	0.8366
No log	48.0	144	0.6384	0.5618	0.6385	0.7991
No log	49.0	147	0.7037	0.5537	0.7039	0.8390
No log	50.0	150	0.7085	0.5607	0.7085	0.8418
No log	51.0	153	0.7524	0.5395	0.7525	0.8675
No log	52.0	156	0.7658	0.5364	0.7660	0.8752
No log	53.0	159	0.7114	0.5445	0.7115	0.8435
No log	54.0	162	0.7088	0.5506	0.7089	0.8420
No log	55.0	165	0.6907	0.5552	0.6908	0.8312
No log	56.0	168	0.8073	0.5125	0.8076	0.8987
No log	57.0	171	0.7215	0.5432	0.7216	0.8495

Framework versions

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