ASAP_FineTuningBERT_UnAugV7_k1_task1_organization_k1_k1_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: 0.5921
• Qwk: 0.5836
• Mse: 0.5917
• Rmse: 0.7693

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	1	12.1901	0.0025	12.1905	3.4915
No log	2.0	2	10.3453	0.0101	10.3455	3.2164
No log	3.0	3	9.7517	-0.0008	9.7519	3.1228
No log	4.0	4	9.1850	-0.0008	9.1852	3.0307
No log	5.0	5	8.7554	-0.0008	8.7554	2.9590
No log	6.0	6	8.0617	0.0	8.0618	2.8393
No log	7.0	7	7.3258	0.0	7.3261	2.7067
No log	8.0	8	6.8259	0.0	6.8262	2.6127
No log	9.0	9	6.4311	0.0	6.4314	2.5360
No log	10.0	10	6.0015	0.0056	6.0018	2.4499
No log	11.0	11	5.4675	0.0256	5.4679	2.3384
No log	12.0	12	4.5924	0.0155	4.5929	2.1431
No log	13.0	13	3.5551	0.0039	3.5554	1.8856
No log	14.0	14	3.6816	0.0	3.6817	1.9188
No log	15.0	15	2.9983	0.0	2.9986	1.7316
No log	16.0	16	2.6228	0.0029	2.6234	1.6197
No log	17.0	17	2.5465	0.0063	2.5471	1.5960
No log	18.0	18	2.1613	0.0995	2.1619	1.4703
No log	19.0	19	1.8649	0.0615	1.8654	1.3658
No log	20.0	20	1.7301	0.0345	1.7305	1.3155
No log	21.0	21	1.5243	0.0280	1.5248	1.2348
No log	22.0	22	1.4077	0.0318	1.4082	1.1867
No log	23.0	23	1.3096	0.0318	1.3101	1.1446
No log	24.0	24	1.1806	0.0213	1.1811	1.0868
No log	25.0	25	1.0827	0.0107	1.0832	1.0408
No log	26.0	26	0.9873	0.0107	0.9878	0.9939
No log	27.0	27	0.9156	0.0107	0.9161	0.9571
No log	28.0	28	0.8542	0.1216	0.8546	0.9245
No log	29.0	29	0.8048	0.3729	0.8052	0.8974
No log	30.0	30	0.7601	0.4521	0.7605	0.8721
No log	31.0	31	0.7144	0.4591	0.7148	0.8454
No log	32.0	32	0.6580	0.4725	0.6583	0.8114
No log	33.0	33	0.6229	0.4623	0.6232	0.7895
No log	34.0	34	0.6088	0.4729	0.6091	0.7805
No log	35.0	35	0.5700	0.4738	0.5703	0.7552
No log	36.0	36	0.5596	0.4700	0.5596	0.7481
No log	37.0	37	0.5604	0.4648	0.5603	0.7485
No log	38.0	38	0.5303	0.4720	0.5303	0.7282
No log	39.0	39	0.4989	0.5027	0.4990	0.7064
No log	40.0	40	0.4797	0.4949	0.4797	0.6926
No log	41.0	41	0.4722	0.4897	0.4722	0.6871
No log	42.0	42	0.4747	0.5093	0.4745	0.6889
No log	43.0	43	0.5165	0.4729	0.5163	0.7185
No log	44.0	44	0.5447	0.4899	0.5444	0.7379
No log	45.0	45	0.5171	0.5805	0.5167	0.7188
No log	46.0	46	0.5004	0.6158	0.5001	0.7072
No log	47.0	47	0.5045	0.6017	0.5041	0.7100
No log	48.0	48	0.5041	0.6133	0.5038	0.7098
No log	49.0	49	0.5205	0.6066	0.5201	0.7212
No log	50.0	50	0.5351	0.6135	0.5346	0.7312
No log	51.0	51	0.5426	0.6153	0.5422	0.7363
No log	52.0	52	0.5463	0.6139	0.5458	0.7388
No log	53.0	53	0.5325	0.6041	0.5321	0.7295
No log	54.0	54	0.5188	0.6113	0.5185	0.7201
No log	55.0	55	0.5289	0.6095	0.5286	0.7271
No log	56.0	56	0.5921	0.5836	0.5917	0.7693

Framework versions

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