fold_3

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: 0.9679
• Unbiased Gwet Ac1: 0.1737
• Unbiased Kripp Alpha: 0.3725
• Qwk: 0.4428
• Mse: 0.9682
• Rmse: 0.9840

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	10.9112	0.0004	-0.8331	0.0012	10.9086	3.3028
No log	2.0	4	9.8463	0.0	-0.8450	0.0	9.8442	3.1376
No log	3.0	6	7.9872	0.0	-0.8450	0.0	7.9857	2.8259
No log	4.0	8	6.5936	0.0	-0.8450	0.0	6.5923	2.5675
No log	5.0	10	5.4361	0.0080	-0.6818	0.0205	5.4350	2.3313
No log	6.0	12	4.5422	0.0062	-0.7140	0.0076	4.5411	2.1310
No log	7.0	14	3.8632	0.0	-0.7309	0.0	3.8621	1.9652
No log	8.0	16	3.3408	0.0	-0.7309	0.0	3.3395	1.8274
No log	9.0	18	2.6894	0.0162	-0.4627	0.0303	2.6886	1.6397
No log	10.0	20	2.4224	0.0352	-0.3567	-0.0119	2.4216	1.5562
No log	11.0	22	1.8935	0.0037	-0.3013	0.0147	1.8928	1.3758
No log	12.0	24	1.7246	0.0009	-0.3029	0.0102	1.7238	1.3129
No log	13.0	26	1.4728	0.0009	-0.3029	0.0102	1.4720	1.2133
No log	14.0	28	1.4406	0.0028	-0.2838	0.0302	1.4400	1.2000
No log	15.0	30	1.4410	0.0037	-0.2743	0.0401	1.4405	1.2002
No log	16.0	32	1.3592	0.0037	-0.2743	0.0401	1.3587	1.1656
No log	17.0	34	1.2802	0.0037	-0.2743	0.0401	1.2797	1.1312
No log	18.0	36	1.1438	0.0028	-0.2838	0.0302	1.1433	1.0693
No log	19.0	38	1.0841	0.0028	-0.2838	0.0302	1.0835	1.0409
No log	20.0	40	1.0108	0.0028	-0.2838	0.0302	1.0103	1.0052
No log	21.0	42	0.9483	0.0292	-0.0108	0.1367	0.9480	0.9737
No log	22.0	44	0.9575	0.0742	0.0796	0.1955	0.9574	0.9785
No log	23.0	46	0.8089	0.1912	0.3440	0.3669	0.8088	0.8993
No log	24.0	48	0.7144	0.1398	0.3876	0.3988	0.7141	0.8450
No log	25.0	50	0.6682	0.1443	0.3907	0.4126	0.6679	0.8173
No log	26.0	52	0.6238	0.2224	0.4450	0.4816	0.6238	0.7898
No log	27.0	54	0.7536	0.2575	0.4292	0.4892	0.7537	0.8682
No log	28.0	56	0.8404	0.1738	0.3979	0.4678	0.8406	0.9168
No log	29.0	58	0.6570	0.2704	0.4635	0.5180	0.6570	0.8106
No log	30.0	60	0.5494	0.2269	0.4504	0.5097	0.5494	0.7412
No log	31.0	62	0.5615	0.2551	0.4790	0.5394	0.5614	0.7493
No log	32.0	64	0.7951	0.2052	0.3721	0.4521	0.7951	0.8917
No log	33.0	66	0.7617	0.2007	0.3813	0.4612	0.7617	0.8727
No log	34.0	68	0.5478	0.2747	0.4955	0.5420	0.5479	0.7402
No log	35.0	70	0.5313	0.2787	0.5023	0.5516	0.5314	0.7290
No log	36.0	72	0.7620	0.2091	0.4196	0.4899	0.7622	0.8731
No log	37.0	74	0.8693	0.1787	0.3826	0.4640	0.8695	0.9325
No log	38.0	76	0.5556	0.2973	0.5413	0.5814	0.5559	0.7456
No log	39.0	78	0.5383	0.2785	0.5101	0.5664	0.5385	0.7338
No log	40.0	80	0.5336	0.2779	0.5325	0.5824	0.5338	0.7306
No log	41.0	82	0.5909	0.2950	0.5230	0.5754	0.5911	0.7688
No log	42.0	84	0.6910	0.2401	0.4750	0.5261	0.6913	0.8315
No log	43.0	86	0.7529	0.2136	0.4463	0.5033	0.7532	0.8679
No log	44.0	88	0.6987	0.2522	0.4928	0.5440	0.6990	0.8361
No log	45.0	90	0.5650	0.2768	0.5486	0.5994	0.5652	0.7518
No log	46.0	92	0.5696	0.2778	0.5538	0.6050	0.5698	0.7548
No log	47.0	94	0.6059	0.2868	0.5338	0.5791	0.6062	0.7786
No log	48.0	96	0.6107	0.2817	0.5207	0.5693	0.6110	0.7817
No log	49.0	98	0.5900	0.2543	0.5192	0.5823	0.5903	0.7683
No log	50.0	100	0.6303	0.2040	0.5091	0.5733	0.6306	0.7941
No log	51.0	102	0.6045	0.2586	0.5216	0.5772	0.6048	0.7777
No log	52.0	104	0.6263	0.2536	0.4975	0.5414	0.6266	0.7916
No log	53.0	106	0.6137	0.2299	0.5097	0.5519	0.6140	0.7836
No log	54.0	108	0.6180	0.2708	0.5341	0.5675	0.6182	0.7863
No log	55.0	110	0.6335	0.2654	0.5176	0.5519	0.6336	0.7960
No log	56.0	112	0.6511	0.2131	0.4970	0.5386	0.6511	0.8069
No log	57.0	114	0.6547	0.2366	0.5167	0.5537	0.6546	0.8091
No log	58.0	116	0.7089	0.2448	0.4935	0.5244	0.7089	0.8420
No log	59.0	118	0.8342	0.1901	0.4026	0.4441	0.8343	0.9134
No log	60.0	120	0.8733	0.1610	0.3750	0.4274	0.8735	0.9346
No log	61.0	122	0.8533	0.1930	0.4014	0.4448	0.8534	0.9238
No log	62.0	124	0.7273	0.2353	0.4673	0.5033	0.7273	0.8528
No log	63.0	126	0.6740	0.2485	0.5159	0.5395	0.6739	0.8209
No log	64.0	128	0.7042	0.2216	0.5027	0.5462	0.7041	0.8391
No log	65.0	130	0.6792	0.2530	0.5274	0.5538	0.6791	0.8241
No log	66.0	132	0.7837	0.2395	0.4682	0.5121	0.7839	0.8854
No log	67.0	134	0.9513	0.1706	0.3735	0.4416	0.9516	0.9755
No log	68.0	136	0.9679	0.1737	0.3725	0.4428	0.9682	0.9840

Framework versions

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