ASAP_FineTuningBERT_AugV14_k3_task1_organization_k3_k3_fold2

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: 0.5951
• Qwk: 0.4634
• Mse: 0.5952
• Rmse: 0.7715

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	6	9.5190	0.0	9.5189	3.0853
No log	2.0	12	5.2330	0.0117	5.2334	2.2877
No log	3.0	18	2.6472	0.0	2.6476	1.6272
No log	4.0	24	1.2049	0.0107	1.2055	1.0979
No log	5.0	30	0.7820	0.3897	0.7823	0.8845
No log	6.0	36	1.0508	0.2206	1.0509	1.0252
No log	7.0	42	1.0634	0.1906	1.0640	1.0315
No log	8.0	48	0.6567	0.3530	0.6571	0.8106
No log	9.0	54	0.8152	0.2837	0.8159	0.9032
No log	10.0	60	0.8988	0.2484	0.8995	0.9484
No log	11.0	66	0.8397	0.3581	0.8402	0.9166
No log	12.0	72	0.7317	0.3678	0.7324	0.8558
No log	13.0	78	0.8232	0.4140	0.8240	0.9078
No log	14.0	84	0.7426	0.4861	0.7429	0.8619
No log	15.0	90	0.8470	0.3187	0.8475	0.9206
No log	16.0	96	0.7448	0.4374	0.7451	0.8632
No log	17.0	102	0.7999	0.3555	0.8005	0.8947
No log	18.0	108	0.7590	0.3685	0.7597	0.8716
No log	19.0	114	0.7169	0.3607	0.7176	0.8471
No log	20.0	120	0.6535	0.4405	0.6539	0.8087
No log	21.0	126	0.7114	0.4319	0.7119	0.8437
No log	22.0	132	0.6795	0.4210	0.6797	0.8245
No log	23.0	138	0.8065	0.4219	0.8065	0.8981
No log	24.0	144	0.7746	0.3643	0.7750	0.8803
No log	25.0	150	0.6531	0.4158	0.6533	0.8082
No log	26.0	156	0.7311	0.3635	0.7313	0.8552
No log	27.0	162	0.6534	0.4619	0.6534	0.8083
No log	28.0	168	0.6185	0.4921	0.6185	0.7865
No log	29.0	174	0.6876	0.4302	0.6875	0.8292
No log	30.0	180	0.6927	0.3907	0.6929	0.8324
No log	31.0	186	0.7951	0.4346	0.7950	0.8917
No log	32.0	192	0.7032	0.3906	0.7035	0.8388
No log	33.0	198	0.8648	0.4186	0.8646	0.9298
No log	34.0	204	0.7001	0.3624	0.7004	0.8369
No log	35.0	210	0.6194	0.4804	0.6194	0.7870
No log	36.0	216	0.6399	0.4496	0.6400	0.8000
No log	37.0	222	0.6305	0.4533	0.6307	0.7942
No log	38.0	228	0.6637	0.4433	0.6637	0.8147
No log	39.0	234	0.6571	0.4383	0.6573	0.8107
No log	40.0	240	0.6261	0.4398	0.6264	0.7915
No log	41.0	246	0.6517	0.4524	0.6519	0.8074
No log	42.0	252	0.6569	0.4738	0.6571	0.8106
No log	43.0	258	0.6042	0.4799	0.6043	0.7773
No log	44.0	264	0.6240	0.4716	0.6242	0.7900
No log	45.0	270	0.6581	0.4327	0.6583	0.8114
No log	46.0	276	0.6090	0.4603	0.6091	0.7804
No log	47.0	282	0.6117	0.4981	0.6117	0.7821
No log	48.0	288	0.5910	0.4620	0.5910	0.7688
No log	49.0	294	0.5999	0.5009	0.6000	0.7746
No log	50.0	300	0.5903	0.4531	0.5903	0.7683
No log	51.0	306	0.6357	0.4559	0.6358	0.7974
No log	52.0	312	0.6007	0.4465	0.6009	0.7752
No log	53.0	318	0.5895	0.4417	0.5896	0.7679
No log	54.0	324	0.6074	0.4883	0.6075	0.7794
No log	55.0	330	0.6068	0.4619	0.6070	0.7791
No log	56.0	336	0.6043	0.4738	0.6044	0.7774
No log	57.0	342	0.6143	0.4709	0.6144	0.7838
No log	58.0	348	0.5919	0.4544	0.5920	0.7694
No log	59.0	354	0.5905	0.4871	0.5905	0.7685
No log	60.0	360	0.6012	0.4630	0.6012	0.7754
No log	61.0	366	0.5914	0.4447	0.5915	0.7691
No log	62.0	372	0.5832	0.4784	0.5834	0.7638
No log	63.0	378	0.5891	0.4828	0.5892	0.7676
No log	64.0	384	0.6170	0.4546	0.6171	0.7855
No log	65.0	390	0.6095	0.4598	0.6095	0.7807
No log	66.0	396	0.5902	0.4829	0.5903	0.7683
No log	67.0	402	0.5815	0.4881	0.5815	0.7626
No log	68.0	408	0.6017	0.4941	0.6018	0.7757
No log	69.0	414	0.5951	0.4634	0.5952	0.7715

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0