ASAP_FineTuningBERT_AugV8_k10_task1_organization_fold4

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: 0.6089
• Qwk: 0.5812
• Mse: 0.6089
• Rmse: 0.7803

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	0.25	2	10.3423	0.0092	10.3423	3.2159
No log	0.5	4	9.0145	0.0037	9.0145	3.0024
No log	0.75	6	7.7197	0.0018	7.7197	2.7784
No log	1.0	8	6.8532	0.0018	6.8532	2.6179
6.4018	1.25	10	6.0474	0.0018	6.0474	2.4591
6.4018	1.5	12	5.1275	0.0128	5.1275	2.2644
6.4018	1.75	14	4.1454	0.0040	4.1454	2.0360
6.4018	2.0	16	2.5905	0.0052	2.5905	1.6095
6.4018	2.25	18	2.1296	0.1216	2.1296	1.4593
2.7798	2.5	20	1.6199	0.0544	1.6199	1.2728
2.7798	2.75	22	1.5879	0.0238	1.5879	1.2601
2.7798	3.0	24	1.2737	0.0238	1.2737	1.1286
2.7798	3.25	26	1.1221	0.0212	1.1221	1.0593
2.7798	3.5	28	1.2226	0.0316	1.2226	1.1057
1.7553	3.75	30	1.5063	0.0420	1.5063	1.2273
1.7553	4.0	32	2.0288	0.1601	2.0288	1.4244
1.7553	4.25	34	1.8433	0.1109	1.8433	1.3577
1.7553	4.5	36	1.5115	0.0757	1.5115	1.2294
1.7553	4.75	38	1.3201	0.0445	1.3201	1.1490
1.7333	5.0	40	1.2861	0.0558	1.2861	1.1341
1.7333	5.25	42	1.1777	0.0509	1.1777	1.0852
1.7333	5.5	44	1.1542	0.0445	1.1542	1.0743
1.7333	5.75	46	1.0644	0.0509	1.0644	1.0317
1.7333	6.0	48	1.8744	0.1643	1.8744	1.3691
1.7201	6.25	50	2.1763	0.1570	2.1763	1.4752
1.7201	6.5	52	1.7457	0.1783	1.7457	1.3213
1.7201	6.75	54	1.1182	0.0610	1.1182	1.0574
1.7201	7.0	56	1.2662	0.0445	1.2662	1.1253
1.7201	7.25	58	0.9480	0.0666	0.9480	0.9737
1.5405	7.5	60	1.0434	0.1008	1.0434	1.0215
1.5405	7.75	62	1.1104	0.0958	1.1104	1.0537
1.5405	8.0	64	0.9350	0.1352	0.9350	0.9669
1.5405	8.25	66	0.8286	0.3014	0.8286	0.9103
1.5405	8.5	68	0.8852	0.2636	0.8852	0.9409
1.4268	8.75	70	0.9632	0.2423	0.9632	0.9814
1.4268	9.0	72	0.8796	0.3042	0.8796	0.9379
1.4268	9.25	74	0.7551	0.4221	0.7551	0.8690
1.4268	9.5	76	0.8424	0.3961	0.8424	0.9178
1.4268	9.75	78	1.2577	0.2541	1.2577	1.1215
1.0054	10.0	80	0.8085	0.3728	0.8085	0.8992
1.0054	10.25	82	0.5684	0.4533	0.5684	0.7540
1.0054	10.5	84	0.5676	0.5215	0.5676	0.7534
1.0054	10.75	86	0.6962	0.4452	0.6962	0.8344
1.0054	11.0	88	0.5826	0.4438	0.5826	0.7633
0.6941	11.25	90	0.6648	0.3708	0.6648	0.8154
0.6941	11.5	92	0.5580	0.4754	0.5580	0.7470
0.6941	11.75	94	0.6176	0.5360	0.6176	0.7859
0.6941	12.0	96	0.5918	0.5448	0.5918	0.7693
0.6941	12.25	98	0.5433	0.5664	0.5433	0.7371
0.4638	12.5	100	0.5908	0.5526	0.5908	0.7686
0.4638	12.75	102	0.5956	0.5379	0.5956	0.7718
0.4638	13.0	104	0.6086	0.5680	0.6086	0.7801
0.4638	13.25	106	0.5973	0.5890	0.5973	0.7728
0.4638	13.5	108	0.5626	0.5852	0.5626	0.7501
0.2885	13.75	110	0.6225	0.5863	0.6225	0.7890
0.2885	14.0	112	0.6296	0.5823	0.6296	0.7934
0.2885	14.25	114	0.5746	0.6220	0.5746	0.7580
0.2885	14.5	116	0.5443	0.6016	0.5443	0.7378
0.2885	14.75	118	0.5433	0.6266	0.5433	0.7371
0.2075	15.0	120	0.5344	0.6296	0.5344	0.7310
0.2075	15.25	122	0.5613	0.6253	0.5613	0.7492
0.2075	15.5	124	0.5944	0.6347	0.5944	0.7710
0.2075	15.75	126	0.6619	0.5735	0.6619	0.8136
0.2075	16.0	128	0.6595	0.5727	0.6595	0.8121
0.176	16.25	130	0.6753	0.5760	0.6753	0.8218
0.176	16.5	132	0.6061	0.5965	0.6061	0.7785
0.176	16.75	134	0.6268	0.5990	0.6268	0.7917
0.176	17.0	136	0.5943	0.5971	0.5943	0.7709
0.176	17.25	138	0.5691	0.6009	0.5691	0.7544
0.17	17.5	140	0.6318	0.6054	0.6318	0.7949
0.17	17.75	142	0.7145	0.5573	0.7145	0.8453
0.17	18.0	144	0.6089	0.5812	0.6089	0.7803

Framework versions

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