Version2ASAP_FineTuningBERT_AugV5_k5_task1_organization_k5_k5_fold1

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: 0.7101
• Qwk: 0.4831
• Mse: 0.7099
• Rmse: 0.8425

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	8	8.1679	0.0	8.1656	2.8575
No log	2.0	16	3.4528	0.0	3.4509	1.8576
No log	3.0	24	1.8350	0.0	1.8333	1.3540
No log	4.0	32	1.3154	0.0482	1.3139	1.1462
No log	5.0	40	1.5742	0.0482	1.5726	1.2540
No log	6.0	48	1.5306	0.1390	1.5293	1.2366
No log	7.0	56	1.5540	0.1537	1.5526	1.2460
No log	8.0	64	2.6429	0.0806	2.6420	1.6254
No log	9.0	72	1.4340	0.2207	1.4338	1.1974
No log	10.0	80	2.3066	0.1509	2.3064	1.5187
No log	11.0	88	1.3460	0.3264	1.3460	1.1602
No log	12.0	96	0.9779	0.3565	0.9780	0.9889
No log	13.0	104	0.9231	0.3877	0.9229	0.9607
No log	14.0	112	0.9351	0.4409	0.9349	0.9669
No log	15.0	120	1.2794	0.3250	1.2792	1.1310
No log	16.0	128	1.2811	0.3293	1.2808	1.1317
No log	17.0	136	1.1820	0.3206	1.1816	1.0870
No log	18.0	144	1.3402	0.3045	1.3397	1.1575
No log	19.0	152	0.9547	0.4685	0.9544	0.9769
No log	20.0	160	0.8900	0.4670	0.8898	0.9433
No log	21.0	168	0.8535	0.4721	0.8536	0.9239
No log	22.0	176	0.7688	0.4794	0.7688	0.8768
No log	23.0	184	0.8412	0.4710	0.8411	0.9171
No log	24.0	192	1.0374	0.4030	1.0373	1.0185
No log	25.0	200	0.8727	0.4720	0.8725	0.9341
No log	26.0	208	0.9580	0.4546	0.9577	0.9786
No log	27.0	216	1.1880	0.3583	1.1876	1.0898
No log	28.0	224	0.7510	0.5134	0.7509	0.8665
No log	29.0	232	0.8038	0.4670	0.8036	0.8965
No log	30.0	240	0.7920	0.4910	0.7918	0.8899
No log	31.0	248	0.7162	0.4906	0.7160	0.8462
No log	32.0	256	0.7490	0.4937	0.7488	0.8654
No log	33.0	264	0.7302	0.4918	0.7299	0.8544
No log	34.0	272	0.7418	0.4965	0.7416	0.8612
No log	35.0	280	0.7991	0.4900	0.7989	0.8938
No log	36.0	288	0.8070	0.4803	0.8067	0.8982
No log	37.0	296	0.8538	0.4725	0.8535	0.9238
No log	38.0	304	0.7971	0.4879	0.7969	0.8927
No log	39.0	312	0.7717	0.5083	0.7716	0.8784
No log	40.0	320	0.7962	0.4848	0.7960	0.8922
No log	41.0	328	0.8147	0.4675	0.8145	0.9025
No log	42.0	336	1.0344	0.4007	1.0341	1.0169
No log	43.0	344	1.0062	0.4225	1.0059	1.0030
No log	44.0	352	0.7815	0.4698	0.7813	0.8839
No log	45.0	360	0.7330	0.4933	0.7327	0.8560
No log	46.0	368	0.7203	0.5140	0.7200	0.8485
No log	47.0	376	0.7162	0.5143	0.7160	0.8462
No log	48.0	384	0.7765	0.4511	0.7762	0.8810
No log	49.0	392	0.7686	0.4676	0.7685	0.8766
No log	50.0	400	0.6906	0.5203	0.6904	0.8309
No log	51.0	408	0.8322	0.4764	0.8320	0.9121
No log	52.0	416	0.9142	0.4477	0.9140	0.9560
No log	53.0	424	0.8925	0.4482	0.8923	0.9446
No log	54.0	432	0.8156	0.4669	0.8153	0.9030
No log	55.0	440	0.7400	0.4877	0.7397	0.8601
No log	56.0	448	0.7052	0.5092	0.7050	0.8396
No log	57.0	456	0.7800	0.4730	0.7798	0.8831
No log	58.0	464	0.8042	0.4678	0.8040	0.8967
No log	59.0	472	0.7523	0.4652	0.7521	0.8672
No log	60.0	480	0.7456	0.4860	0.7454	0.8634
No log	61.0	488	0.8170	0.4646	0.8167	0.9037
No log	62.0	496	0.8436	0.4635	0.8433	0.9183
0.5289	63.0	504	0.7795	0.4793	0.7793	0.8828
0.5289	64.0	512	0.8339	0.4525	0.8336	0.9130
0.5289	65.0	520	0.7139	0.4887	0.7137	0.8448
0.5289	66.0	528	0.7020	0.5005	0.7018	0.8377
0.5289	67.0	536	0.7646	0.4682	0.7643	0.8742
0.5289	68.0	544	0.8408	0.4717	0.8404	0.9167
0.5289	69.0	552	0.7478	0.4691	0.7476	0.8646
0.5289	70.0	560	0.7101	0.4831	0.7099	0.8425

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0