ASAP_FineTuningBERT_AugV8_k4_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: 0.7658
• Qwk: 0.5782
• Mse: 0.7658
• Rmse: 0.8751

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	10.6551	-0.0055	10.6551	3.2642
No log	1.0	4	9.3998	0.0018	9.3998	3.0659
No log	1.5	6	8.5348	0.0018	8.5348	2.9214
No log	2.0	8	7.6490	0.0018	7.6490	2.7657
7.3106	2.5	10	6.9636	0.0	6.9636	2.6389
7.3106	3.0	12	6.0798	0.0	6.0798	2.4657
7.3106	3.5	14	5.3738	-0.0001	5.3738	2.3181
7.3106	4.0	16	4.3776	0.0079	4.3776	2.0923
7.3106	4.5	18	3.6539	0.0040	3.6539	1.9115
3.6722	5.0	20	3.1273	0.0040	3.1273	1.7684
3.6722	5.5	22	2.4647	0.0	2.4647	1.5699
3.6722	6.0	24	2.1820	0.0962	2.1820	1.4772
3.6722	6.5	26	1.7263	0.0239	1.7263	1.3139
3.6722	7.0	28	1.4220	0.0213	1.4220	1.1925
2.1624	7.5	30	1.2235	0.0213	1.2235	1.1061
2.1624	8.0	32	1.1225	0.0213	1.1225	1.0595
2.1624	8.5	34	1.2824	0.0317	1.2824	1.1324
2.1624	9.0	36	1.2670	0.0626	1.2670	1.1256
2.1624	9.5	38	1.4089	0.1007	1.4089	1.1870
1.7529	10.0	40	1.6309	0.1979	1.6309	1.2771
1.7529	10.5	42	1.2237	0.1450	1.2237	1.1062
1.7529	11.0	44	1.1384	0.1663	1.1384	1.0670
1.7529	11.5	46	1.4952	0.2174	1.4952	1.2228
1.7529	12.0	48	1.2157	0.2620	1.2157	1.1026
1.5899	12.5	50	0.8029	0.4588	0.8029	0.8961
1.5899	13.0	52	1.0831	0.3626	1.0831	1.0407
1.5899	13.5	54	0.7906	0.4625	0.7906	0.8892
1.5899	14.0	56	0.7230	0.3870	0.7230	0.8503
1.5899	14.5	58	0.6752	0.4900	0.6752	0.8217
1.1781	15.0	60	1.2732	0.3767	1.2732	1.1284
1.1781	15.5	62	0.8177	0.5053	0.8177	0.9043
1.1781	16.0	64	0.6513	0.4603	0.6513	0.8070
1.1781	16.5	66	0.6487	0.4778	0.6487	0.8054
1.1781	17.0	68	1.0368	0.4193	1.0368	1.0182
0.919	17.5	70	0.6965	0.4902	0.6965	0.8346
0.919	18.0	72	0.7103	0.4844	0.7103	0.8428
0.919	18.5	74	0.6683	0.5540	0.6683	0.8175
0.919	19.0	76	0.7479	0.5456	0.7479	0.8648
0.919	19.5	78	0.6432	0.5743	0.6432	0.8020
0.5635	20.0	80	0.6621	0.5585	0.6621	0.8137
0.5635	20.5	82	0.7227	0.5564	0.7227	0.8501
0.5635	21.0	84	0.6786	0.5778	0.6786	0.8238
0.5635	21.5	86	0.6798	0.5482	0.6798	0.8245
0.5635	22.0	88	0.6770	0.6057	0.6770	0.8228
0.2963	22.5	90	0.6661	0.6067	0.6661	0.8161
0.2963	23.0	92	0.6876	0.5485	0.6876	0.8292
0.2963	23.5	94	0.6650	0.5927	0.6650	0.8155
0.2963	24.0	96	0.6599	0.5889	0.6599	0.8123
0.2963	24.5	98	0.6240	0.6102	0.6240	0.7899
0.256	25.0	100	0.6041	0.6219	0.6041	0.7772
0.256	25.5	102	0.5972	0.6155	0.5972	0.7728
0.256	26.0	104	0.5639	0.6407	0.5639	0.7509
0.256	26.5	106	0.5869	0.6271	0.5869	0.7661
0.256	27.0	108	0.9721	0.4927	0.9721	0.9859
0.198	27.5	110	0.6480	0.5891	0.6480	0.8050
0.198	28.0	112	0.6183	0.6164	0.6183	0.7863
0.198	28.5	114	0.5563	0.6325	0.5563	0.7458
0.198	29.0	116	0.6943	0.5662	0.6943	0.8333
0.198	29.5	118	0.6054	0.6342	0.6054	0.7781
0.215	30.0	120	0.8387	0.5590	0.8387	0.9158
0.215	30.5	122	0.7214	0.5938	0.7214	0.8494
0.215	31.0	124	0.7658	0.5782	0.7658	0.8751

Framework versions

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