fold_1

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: 1.7100
• Unbiased Gwet Ac1: 0.0591
• Unbiased Kripp Alpha: -0.0282
• Qwk: 0.0425
• Mse: 1.7082
• Rmse: 1.3070

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 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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Unbiased Gwet Ac1	Unbiased Kripp Alpha	Qwk	Mse	Rmse
No log	1.0	2	10.8915	-0.0076	-0.7285	0.0179	10.8884	3.2998
No log	2.0	4	7.3303	0.0	-0.8443	0.0	7.3281	2.7070
No log	3.0	6	6.2423	0.0	-0.8443	0.0	6.2400	2.4980
No log	4.0	8	5.7437	0.0122	-0.7745	-0.0336	5.7415	2.3961
No log	5.0	10	5.0574	0.0	-0.7268	0.0	5.0552	2.2484
No log	6.0	12	4.4212	0.0	-0.7268	0.0	4.4191	2.1022
No log	7.0	14	4.2088	0.0	-0.7268	0.0	4.2068	2.0510
No log	8.0	16	3.5900	0.0	-0.7268	0.0	3.5881	1.8942
No log	9.0	18	3.0096	0.0	-0.7268	0.0	3.0076	1.7342
No log	10.0	20	2.5027	0.0	-0.7268	0.0	2.5009	1.5814
No log	11.0	22	2.0334	0.0291	-0.1314	0.0957	2.0316	1.4253
No log	12.0	24	1.7063	0.0235	-0.1427	0.0899	1.7046	1.3056
No log	13.0	26	1.5234	0.0177	-0.1791	0.0645	1.5219	1.2337
No log	14.0	28	1.5443	0.0165	-0.1890	0.0545	1.5429	1.2421
No log	15.0	30	1.1074	0.0177	-0.1791	0.0645	1.1060	1.0517
No log	16.0	32	1.1227	0.0177	-0.1791	0.0645	1.1213	1.0589
No log	17.0	34	0.8893	0.0603	0.1916	0.1967	0.8881	0.9424
No log	18.0	36	1.1833	-0.0934	-0.3027	-0.1458	1.1820	1.0872
No log	19.0	38	1.2590	-0.0713	-0.3140	-0.1316	1.2576	1.1214
No log	20.0	40	0.8762	-0.0115	0.0242	0.1128	0.8752	0.9355
No log	21.0	42	0.8402	0.1462	0.2549	0.2978	0.8389	0.9159
No log	22.0	44	0.9614	0.0753	0.0261	0.1933	0.9599	0.9798
No log	23.0	46	0.9663	0.0703	0.0305	0.2006	0.9648	0.9822
No log	24.0	48	0.6612	0.1609	0.4418	0.4523	0.6601	0.8124
No log	25.0	50	0.6863	0.0904	0.2893	0.3795	0.6856	0.8280
No log	26.0	52	0.6445	0.1006	0.3049	0.4069	0.6437	0.8023
No log	27.0	54	0.5540	0.1808	0.4119	0.4941	0.5531	0.7437
No log	28.0	56	0.5439	0.2765	0.5587	0.5732	0.5429	0.7368
No log	29.0	58	0.9548	0.1684	0.3407	0.4440	0.9535	0.9765
No log	30.0	60	0.7077	0.2661	0.5353	0.5733	0.7065	0.8406
No log	31.0	62	0.5210	0.2015	0.4444	0.5164	0.5203	0.7213
No log	32.0	64	0.5295	0.1812	0.4253	0.4922	0.5287	0.7271
No log	33.0	66	0.5641	0.2798	0.5255	0.5482	0.5632	0.7504
No log	34.0	68	1.0915	0.0778	0.1774	0.3259	1.0899	1.0440
No log	35.0	70	0.6484	0.2691	0.5017	0.5240	0.6474	0.8046
No log	36.0	72	0.6547	0.2406	0.4895	0.5198	0.6537	0.8085
No log	37.0	74	1.6375	0.0208	0.0213	0.2301	1.6357	1.2789
No log	38.0	76	1.7421	0.0072	-0.0190	0.1991	1.7401	1.3191
No log	39.0	78	1.0599	0.1110	0.2301	0.3167	1.0585	1.0288
No log	40.0	80	0.7534	0.1786	0.3713	0.4117	0.7523	0.8673
No log	41.0	82	0.8205	0.1488	0.3097	0.3548	0.8194	0.9052
No log	42.0	84	1.1338	0.0678	0.1985	0.2827	1.1323	1.0641
No log	43.0	86	1.2388	0.0719	0.1103	0.2200	1.2372	1.1123
No log	44.0	88	0.8093	0.1132	0.2959	0.3527	0.8084	0.8991
No log	45.0	90	0.8057	0.1062	0.2952	0.3631	0.8049	0.8971
No log	46.0	92	1.0124	0.0730	0.1542	0.2288	1.0110	1.0055
No log	47.0	94	1.5629	0.0810	-0.0076	0.1562	1.5610	1.2494
No log	48.0	96	1.5855	0.0832	-0.0055	0.1572	1.5836	1.2584
No log	49.0	98	1.1875	0.0837	0.1227	0.2147	1.1860	1.0890
No log	50.0	100	0.9652	0.0875	0.2460	0.3200	0.9642	0.9819
No log	51.0	102	0.9583	0.1133	0.2615	0.3269	0.9574	0.9785
No log	52.0	104	0.8415	0.0861	0.3031	0.3630	0.8409	0.9170
No log	53.0	106	0.8424	0.0763	0.2911	0.3466	0.8418	0.9175
No log	54.0	108	0.8857	0.0790	0.2826	0.3195	0.8849	0.9407
No log	55.0	110	1.2987	0.0468	0.0430	0.1428	1.2971	1.1389
No log	56.0	112	1.5208	0.0640	-0.0532	0.0941	1.5189	1.2325
No log	57.0	114	1.2407	0.0343	-0.0006	0.1061	1.2391	1.1131
No log	58.0	116	0.9848	0.0440	0.0882	0.1198	0.9834	0.9917
No log	59.0	118	0.9037	-0.0001	0.1169	0.1476	0.9026	0.9500
No log	60.0	120	1.0132	-0.0246	0.0849	0.1295	1.0122	1.0061
No log	61.0	122	1.1968	-0.0100	0.0831	0.1093	1.1957	1.0935
No log	62.0	124	1.6235	0.0293	-0.0217	0.0333	1.6218	1.2735
No log	63.0	126	1.7100	0.0591	-0.0282	0.0425	1.7082	1.3070

Framework versions

• Transformers 4.51.1
• Pytorch 2.5.1+cu124
• Datasets 3.5.0
• Tokenizers 0.21.0