ASAP_FineTuningBERT_AugV8_k4_task1_organization_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: 1.0202
• Qwk: 0.3753
• Mse: 1.0207
• Rmse: 1.0103

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	0.5	2	10.9512	0.0065	10.9509	3.3092
No log	1.0	4	8.2330	0.0	8.2328	2.8693
No log	1.5	6	7.0535	0.0	7.0532	2.6558
No log	2.0	8	6.0220	0.0062	6.0218	2.4539
6.5395	2.5	10	5.1173	0.0039	5.1171	2.2621
6.5395	3.0	12	4.1790	0.0	4.1789	2.0442
6.5395	3.5	14	3.4673	0.0	3.4672	1.8620
6.5395	4.0	16	2.8669	0.0	2.8668	1.6932
6.5395	4.5	18	2.2884	0.1275	2.2885	1.5128
2.9763	5.0	20	1.8749	0.0345	1.8749	1.3693
2.9763	5.5	22	1.5676	0.0280	1.5677	1.2521
2.9763	6.0	24	1.3053	0.0280	1.3055	1.1426
2.9763	6.5	26	1.1773	0.0345	1.1774	1.0851
2.9763	7.0	28	1.5023	0.1409	1.5023	1.2257
1.9243	7.5	30	1.4926	0.1734	1.4927	1.2217
1.9243	8.0	32	1.9105	0.1467	1.9105	1.3822
1.9243	8.5	34	1.0072	0.0850	1.0073	1.0037
1.9243	9.0	36	0.8196	0.1519	0.8200	0.9055
1.9243	9.5	38	0.8379	0.2106	0.8381	0.9155
1.7441	10.0	40	1.3793	0.1870	1.3795	1.1745
1.7441	10.5	42	1.3691	0.2459	1.3691	1.1701
1.7441	11.0	44	0.8412	0.2989	0.8413	0.9172
1.7441	11.5	46	0.8443	0.3184	0.8443	0.9189
1.7441	12.0	48	1.1500	0.3135	1.1499	1.0723
1.3171	12.5	50	0.9478	0.3527	0.9478	0.9735
1.3171	13.0	52	0.9078	0.3748	0.9080	0.9529
1.3171	13.5	54	1.0723	0.3308	1.0723	1.0355
1.3171	14.0	56	0.7517	0.3727	0.7517	0.8670
1.3171	14.5	58	0.8229	0.3334	0.8230	0.9072
1.0873	15.0	60	0.8219	0.3678	0.8220	0.9066
1.0873	15.5	62	0.9343	0.3402	0.9342	0.9665
1.0873	16.0	64	0.9035	0.3279	0.9034	0.9505
1.0873	16.5	66	0.8606	0.3139	0.8605	0.9276
1.0873	17.0	68	0.8057	0.3566	0.8056	0.8975
0.6977	17.5	70	0.9853	0.3620	0.9853	0.9926
0.6977	18.0	72	0.6940	0.4122	0.6941	0.8331
0.6977	18.5	74	0.7399	0.4333	0.7400	0.8602
0.6977	19.0	76	1.1491	0.3224	1.1493	1.0720
0.6977	19.5	78	1.2577	0.3257	1.2577	1.1215
0.6162	20.0	80	0.8904	0.3918	0.8903	0.9436
0.6162	20.5	82	0.9218	0.3548	0.9216	0.9600
0.6162	21.0	84	1.0861	0.3647	1.0859	1.0421
0.6162	21.5	86	1.1339	0.3778	1.1337	1.0648
0.6162	22.0	88	0.9433	0.3840	0.9430	0.9711
0.3943	22.5	90	1.0313	0.4106	1.0312	1.0155
0.3943	23.0	92	1.2774	0.3575	1.2774	1.1302
0.3943	23.5	94	0.9194	0.4369	0.9195	0.9589
0.3943	24.0	96	0.8937	0.4385	0.8937	0.9454
0.3943	24.5	98	1.3253	0.3563	1.3255	1.1513
0.3118	25.0	100	1.3241	0.3356	1.3243	1.1508
0.3118	25.5	102	0.9723	0.3981	0.9724	0.9861
0.3118	26.0	104	1.0191	0.3828	1.0191	1.0095
0.3118	26.5	106	1.2886	0.3253	1.2886	1.1352
0.3118	27.0	108	1.4195	0.2951	1.4195	1.1914
0.3414	27.5	110	0.9384	0.4089	0.9388	0.9689
0.3414	28.0	112	0.9058	0.4047	0.9061	0.9519
0.3414	28.5	114	1.0059	0.3756	1.0064	1.0032
0.3414	29.0	116	1.0202	0.3753	1.0207	1.0103

Framework versions

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