fold_2

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: 0.6860
• Unbiased Gwet Ac1: 0.1645
• Unbiased Kripp Alpha: 0.3978
• Qwk: 0.4826
• Mse: 0.6862
• Rmse: 0.8283

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	13.8667	0.0	-0.8682	0.0	13.8669	3.7238
No log	2.0	4	11.2585	-0.0077	-0.7897	-0.0019	11.2588	3.3554
No log	3.0	6	9.9494	0.0	-0.8493	0.0	9.9495	3.1543
No log	4.0	8	9.4986	0.0	-0.8493	0.0	9.4988	3.0820
No log	5.0	10	6.6041	-0.0037	-0.7515	0.0085	6.6049	2.5700
No log	6.0	12	6.6811	0.0144	-0.8137	-0.0241	6.6806	2.5847
No log	7.0	14	4.7936	0.0051	-0.6906	0.0211	4.7937	2.1894
No log	8.0	16	3.6890	0.0031	-0.7284	0.0039	3.6897	1.9209
No log	9.0	18	2.9625	0.0031	-0.7284	0.0039	2.9629	1.7213
No log	10.0	20	2.2825	0.0375	-0.2708	0.0713	2.2830	1.5110
No log	11.0	22	1.7107	-0.0023	-0.2679	0.0361	1.7113	1.3082
No log	12.0	24	1.5328	0.0077	-0.2822	0.0268	1.5334	1.2383
No log	13.0	26	1.1126	0.0019	-0.2885	0.0213	1.1130	1.0550
No log	14.0	28	0.9343	0.0099	-0.2528	0.0286	0.9347	0.9668
No log	15.0	30	1.0945	0.0403	-0.1429	0.0703	1.0951	1.0465
No log	16.0	32	0.8949	0.0910	0.0875	0.1570	0.8954	0.9463
No log	17.0	34	0.7196	0.0847	0.2800	0.2880	0.7199	0.8485
No log	18.0	36	0.6785	0.1669	0.4253	0.3972	0.6788	0.8239
No log	19.0	38	0.9353	0.0834	0.0399	0.1830	0.9357	0.9673
No log	20.0	40	0.6663	0.2074	0.4600	0.4454	0.6665	0.8164
No log	21.0	42	0.6293	0.1670	0.4101	0.4102	0.6293	0.7933
No log	22.0	44	0.7201	0.1480	0.3191	0.3502	0.7201	0.8486
No log	23.0	46	0.7950	0.1589	0.2945	0.3688	0.7950	0.8916
No log	24.0	48	0.8493	0.1760	0.4220	0.4799	0.8494	0.9216
No log	25.0	50	0.6508	0.2350	0.5358	0.5766	0.6508	0.8067
No log	26.0	52	0.6383	0.2318	0.4965	0.5382	0.6385	0.7990
No log	27.0	54	0.7376	0.1910	0.3921	0.4543	0.7379	0.8590
No log	28.0	56	0.4943	0.2212	0.4891	0.4950	0.4943	0.7031
No log	29.0	58	0.5398	0.2343	0.5046	0.4943	0.5400	0.7348
No log	30.0	60	1.1495	0.0557	-0.1583	0.1291	1.1499	1.0723
No log	31.0	62	1.3284	0.0366	-0.1900	0.1163	1.3289	1.1528
No log	32.0	64	1.1032	0.0597	-0.1612	0.1255	1.1036	1.0505
No log	33.0	66	0.6629	0.1659	0.3846	0.4176	0.6633	0.8144
No log	34.0	68	0.5792	0.1300	0.3775	0.4239	0.5791	0.7610
No log	35.0	70	0.5951	0.2046	0.4271	0.4911	0.5949	0.7713
No log	36.0	72	0.5759	0.2207	0.4681	0.4985	0.5760	0.7589
No log	37.0	74	1.4122	0.0350	0.0482	0.2925	1.4125	1.1885
No log	38.0	76	1.4282	0.0338	0.0383	0.2883	1.4284	1.1952
No log	39.0	78	0.8152	0.1789	0.3651	0.4589	0.8153	0.9029
No log	40.0	80	0.5022	0.2409	0.5870	0.6120	0.5020	0.7085
No log	41.0	82	0.5191	0.2445	0.5527	0.5812	0.5190	0.7204
No log	42.0	84	0.9344	0.1513	0.2424	0.3849	0.9346	0.9667
No log	43.0	86	1.1275	0.0678	0.0851	0.3026	1.1277	1.0619
No log	44.0	88	0.8308	0.1440	0.3246	0.4292	0.8309	0.9115
No log	45.0	90	0.4983	0.2713	0.5618	0.5826	0.4982	0.7059
No log	46.0	92	0.5774	0.2181	0.4987	0.5680	0.5771	0.7597
No log	47.0	94	0.5246	0.2205	0.5210	0.5722	0.5245	0.7242
No log	48.0	96	0.7050	0.1544	0.3860	0.4484	0.7051	0.8397
No log	49.0	98	1.2352	0.0598	0.0348	0.2697	1.2352	1.1114
No log	50.0	100	1.4866	0.0042	-0.0368	0.2387	1.4867	1.2193
No log	51.0	102	1.3744	0.0308	0.0093	0.2606	1.3744	1.1723
No log	52.0	104	0.9605	0.1383	0.2746	0.3969	0.9604	0.9800
No log	53.0	106	0.5690	0.2417	0.5454	0.5844	0.5688	0.7542
No log	54.0	108	0.6415	0.1839	0.4684	0.5420	0.6413	0.8008
No log	55.0	110	0.6878	0.1446	0.4398	0.5261	0.6876	0.8292
No log	56.0	112	0.5938	0.2095	0.5135	0.5669	0.5936	0.7704
No log	57.0	114	0.8847	0.1548	0.2842	0.3888	0.8845	0.9405
No log	58.0	116	1.2858	0.0710	0.0533	0.2848	1.2858	1.1339
No log	59.0	118	1.1605	0.0934	0.1134	0.3070	1.1605	1.0773
No log	60.0	120	0.7612	0.1882	0.3337	0.4131	0.7612	0.8725
No log	61.0	122	0.6327	0.2089	0.3985	0.4643	0.6327	0.7954
No log	62.0	124	0.6697	0.1903	0.3661	0.4368	0.6697	0.8183
No log	63.0	126	0.7832	0.1741	0.3306	0.4135	0.7833	0.8850
No log	64.0	128	0.8755	0.1347	0.2540	0.3681	0.8756	0.9357
No log	65.0	130	0.8553	0.1380	0.2681	0.3662	0.8554	0.9249
No log	66.0	132	0.7031	0.1933	0.3778	0.4450	0.7032	0.8386
No log	67.0	134	0.6951	0.1784	0.3883	0.4630	0.6953	0.8338
No log	68.0	136	0.7833	0.1635	0.3269	0.4090	0.7835	0.8851
No log	69.0	138	1.1226	0.0895	0.1364	0.3031	1.1227	1.0596
No log	70.0	140	1.4026	0.0514	-0.0342	0.2167	1.4028	1.1844
No log	71.0	142	1.3493	0.0559	-0.0015	0.2321	1.3494	1.1616
No log	72.0	144	1.0388	0.1075	0.1767	0.3192	1.0389	1.0193
No log	73.0	146	0.7402	0.1650	0.3715	0.4393	0.7403	0.8604
No log	74.0	148	0.6865	0.1610	0.3686	0.4522	0.6866	0.8286
No log	75.0	150	0.6860	0.1645	0.3978	0.4826	0.6862	0.8283

Framework versions

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