fold_4

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: 0.6309
• Unbiased Gwet Ac1: 0.2892
• Unbiased Kripp Alpha: 0.5154
• Qwk: 0.5442
• Mse: 0.6309
• Rmse: 0.7943

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	11.0264	-0.0118	-0.7274	0.0206	11.0264	3.3206
No log	2.0	4	10.1682	-0.0050	-0.8152	0.0054	10.1682	3.1888
No log	3.0	6	9.2476	-0.0041	-0.8250	-0.0047	9.2476	3.0410
No log	4.0	8	8.4023	-0.0054	-0.8237	0.0042	8.4023	2.8987
No log	5.0	10	7.3372	-0.0028	-0.8349	0.0018	7.3372	2.7087
No log	6.0	12	6.7718	-0.0028	-0.8349	0.0018	6.7718	2.6023
No log	7.0	14	5.8404	-0.0119	-0.6409	0.0317	5.8404	2.4167
No log	8.0	16	5.2706	-0.0131	-0.6048	0.0592	5.2706	2.2958
No log	9.0	18	4.4730	0.0007	-0.6904	0.0214	4.4730	2.1149
No log	10.0	20	3.9152	0.0033	-0.7112	0.0118	3.9152	1.9787
No log	11.0	22	3.3715	0.0033	-0.7112	0.0118	3.3715	1.8362
No log	12.0	24	3.0821	0.0033	-0.7112	0.0118	3.0821	1.7556
No log	13.0	26	2.5599	0.0002	-0.7196	0.0079	2.5599	1.6000
No log	14.0	28	2.2373	0.0088	-0.1498	0.1313	2.2373	1.4958
No log	15.0	30	1.9770	0.0131	-0.2052	0.0803	1.9770	1.4061
No log	16.0	32	1.6936	0.0042	-0.2237	0.0672	1.6936	1.3014
No log	17.0	34	1.4444	0.0049	-0.2469	0.0445	1.4444	1.2018
No log	18.0	36	1.3404	0.0079	-0.2507	0.0420	1.3404	1.1578
No log	19.0	38	1.1688	0.0079	-0.2507	0.0420	1.1688	1.0811
No log	20.0	40	1.0322	0.0072	-0.2273	0.0648	1.0322	1.0160
No log	21.0	42	0.9298	0.0237	-0.1740	0.1033	0.9298	0.9643
No log	22.0	44	0.8616	0.0035	-0.1024	0.0910	0.8616	0.9282
No log	23.0	46	0.8940	-0.0339	-0.1872	0.0464	0.8940	0.9455
No log	24.0	48	0.8656	0.0349	-0.0191	0.1831	0.8656	0.9304
No log	25.0	50	0.9620	0.0887	0.0625	0.2636	0.9620	0.9808
No log	26.0	52	1.1317	0.0249	-0.0229	0.2346	1.1317	1.0638
No log	27.0	54	0.9522	0.1264	0.1760	0.3204	0.9522	0.9758
No log	28.0	56	0.7545	0.1640	0.4026	0.4734	0.7545	0.8686
No log	29.0	58	0.6778	0.1670	0.4616	0.5227	0.6778	0.8233
No log	30.0	60	0.7054	0.2387	0.5387	0.5655	0.7054	0.8399
No log	31.0	62	0.6837	0.2563	0.5670	0.5805	0.6837	0.8269
No log	32.0	64	0.5985	0.2240	0.5219	0.5537	0.5985	0.7736
No log	33.0	66	0.5619	0.2269	0.5256	0.5588	0.5619	0.7496
No log	34.0	68	0.5762	0.2730	0.5975	0.6053	0.5762	0.7590
No log	35.0	70	0.5503	0.2015	0.4777	0.5354	0.5503	0.7418
No log	36.0	72	0.6658	0.1201	0.3319	0.4158	0.6658	0.8159
No log	37.0	74	0.6725	0.1349	0.3599	0.4580	0.6725	0.8200
No log	38.0	76	0.6004	0.2957	0.5700	0.6031	0.6004	0.7749
No log	39.0	78	0.6630	0.2664	0.6158	0.6211	0.6630	0.8142
No log	40.0	80	0.5848	0.2870	0.5609	0.6024	0.5848	0.7647
No log	41.0	82	0.6953	0.1515	0.3126	0.4273	0.6953	0.8338
No log	42.0	84	0.7533	0.1282	0.2343	0.3531	0.7533	0.8679
No log	43.0	86	0.7389	0.1344	0.2823	0.3966	0.7389	0.8596
No log	44.0	88	0.7110	0.2096	0.3626	0.4500	0.7110	0.8432
No log	45.0	90	0.6533	0.2419	0.4436	0.5091	0.6533	0.8083
No log	46.0	92	0.6370	0.2765	0.5168	0.5589	0.6370	0.7982
No log	47.0	94	0.6482	0.2685	0.4708	0.5201	0.6482	0.8051
No log	48.0	96	0.6309	0.2674	0.4736	0.5137	0.6309	0.7943
No log	49.0	98	0.6205	0.2524	0.4705	0.5083	0.6205	0.7877
No log	50.0	100	0.6709	0.2175	0.4153	0.4780	0.6709	0.8191
No log	51.0	102	0.7670	0.1384	0.3039	0.3879	0.7670	0.8758
No log	52.0	104	0.7444	0.1406	0.3151	0.3900	0.7444	0.8628
No log	53.0	106	0.6048	0.2582	0.4956	0.5264	0.6048	0.7777
No log	54.0	108	0.6071	0.2673	0.4896	0.5159	0.6071	0.7792
No log	55.0	110	0.7582	0.2000	0.3423	0.3929	0.7582	0.8708
No log	56.0	112	0.7947	0.1649	0.3023	0.3609	0.7947	0.8914
No log	57.0	114	0.6972	0.2129	0.4037	0.4554	0.6972	0.8350
No log	58.0	116	0.6060	0.2646	0.5028	0.5436	0.6060	0.7785
No log	59.0	118	0.6440	0.2679	0.4897	0.5308	0.6440	0.8025
No log	60.0	120	0.7145	0.2334	0.4110	0.4617	0.7145	0.8453
No log	61.0	122	0.7068	0.1886	0.3854	0.4419	0.7068	0.8407
No log	62.0	124	0.6725	0.2689	0.4765	0.5173	0.6725	0.8201
No log	63.0	126	0.6400	0.2641	0.4897	0.5225	0.6400	0.8000
No log	64.0	128	0.6351	0.2554	0.5119	0.5568	0.6351	0.7969
No log	65.0	130	0.6172	0.2641	0.5133	0.5458	0.6172	0.7856
No log	66.0	132	0.6485	0.2789	0.4989	0.5343	0.6485	0.8053
No log	67.0	134	0.6669	0.2856	0.4948	0.5267	0.6669	0.8166
No log	68.0	136	0.6416	0.2990	0.5013	0.5342	0.6416	0.8010
No log	69.0	138	0.6093	0.2822	0.5551	0.5812	0.6093	0.7806
No log	70.0	140	0.6105	0.2753	0.5610	0.5865	0.6105	0.7813
No log	71.0	142	0.6137	0.2718	0.5603	0.5859	0.6137	0.7834
No log	72.0	144	0.6250	0.2748	0.5229	0.5533	0.6250	0.7906
No log	73.0	146	0.6977	0.2792	0.5173	0.5407	0.6977	0.8353
No log	74.0	148	0.7680	0.2497	0.4894	0.5253	0.7680	0.8764
No log	75.0	150	0.7736	0.2425	0.4873	0.5182	0.7736	0.8795
No log	76.0	152	0.7542	0.2501	0.4637	0.5039	0.7542	0.8684
No log	77.0	154	0.7603	0.2416	0.4272	0.4841	0.7603	0.8720
No log	78.0	156	0.7630	0.1946	0.3640	0.4356	0.7630	0.8735
No log	79.0	158	0.7593	0.1932	0.3634	0.4379	0.7593	0.8714
No log	80.0	160	0.7343	0.2040	0.3922	0.4571	0.7343	0.8569
No log	81.0	162	0.6865	0.2471	0.4510	0.4961	0.6865	0.8286
No log	82.0	164	0.6591	0.2886	0.5035	0.5341	0.6591	0.8118
No log	83.0	166	0.6387	0.2832	0.5008	0.5321	0.6387	0.7992
No log	84.0	168	0.6222	0.2831	0.5292	0.5582	0.6222	0.7888
No log	85.0	170	0.6247	0.2696	0.5177	0.5444	0.6247	0.7904
No log	86.0	172	0.6352	0.2555	0.4867	0.5291	0.6352	0.7970
No log	87.0	174	0.6533	0.2406	0.4609	0.5124	0.6533	0.8083
No log	88.0	176	0.6746	0.2385	0.4382	0.4919	0.6746	0.8213
No log	89.0	178	0.6812	0.2348	0.4387	0.4912	0.6812	0.8254
No log	90.0	180	0.6834	0.2335	0.4258	0.4818	0.6834	0.8267
No log	91.0	182	0.6777	0.2381	0.4299	0.4846	0.6777	0.8232
No log	92.0	184	0.6765	0.2369	0.4348	0.4898	0.6765	0.8225
No log	93.0	186	0.6695	0.2380	0.4401	0.4926	0.6695	0.8182
No log	94.0	188	0.6532	0.2599	0.4671	0.5100	0.6532	0.8082
No log	95.0	190	0.6379	0.2740	0.5055	0.5373	0.6379	0.7987
No log	96.0	192	0.6304	0.2867	0.5142	0.5435	0.6304	0.7940
No log	97.0	194	0.6276	0.2924	0.5227	0.5530	0.6276	0.7922
No log	98.0	196	0.6309	0.2892	0.5154	0.5442	0.6309	0.7943

Framework versions

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