ASAP_FineTuningBERT_AugV8_k4_task1_organization_fold3

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: 0.7723
• Qwk: 0.4767
• Mse: 0.7721
• Rmse: 0.8787

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	11.5693	-0.0213	11.5681	3.4012
No log	1.0	4	10.1149	0.0	10.1140	3.1803
No log	1.5	6	7.8912	0.0	7.8907	2.8090
No log	2.0	8	5.8536	0.0571	5.8533	2.4194
6.413	2.5	10	4.2944	0.0076	4.2943	2.0723
6.413	3.0	12	3.2357	0.0	3.2358	1.7988
6.413	3.5	14	2.3676	0.1381	2.3679	1.5388
6.413	4.0	16	1.8817	0.0266	1.8821	1.3719
6.413	4.5	18	1.5299	0.0266	1.5305	1.2371
2.22	5.0	20	1.4890	0.0365	1.4896	1.2205
2.22	5.5	22	1.4290	0.0365	1.4297	1.1957
2.22	6.0	24	1.1449	0.0202	1.1456	1.0703
2.22	6.5	26	1.1232	0.0202	1.1240	1.0602
2.22	7.0	28	1.1564	0.0202	1.1571	1.0757
1.7944	7.5	30	1.4511	0.0753	1.4517	1.2048
1.7944	8.0	32	1.8932	0.1640	1.8937	1.3761
1.7944	8.5	34	1.4482	0.1517	1.4486	1.2036
1.7944	9.0	36	1.1352	0.1185	1.1356	1.0657
1.7944	9.5	38	1.0596	0.1558	1.0600	1.0296
1.6569	10.0	40	1.5399	0.2050	1.5402	1.2410
1.6569	10.5	42	1.6531	0.1945	1.6534	1.2858
1.6569	11.0	44	1.0243	0.3529	1.0245	1.0122
1.6569	11.5	46	0.7941	0.3171	0.7946	0.8914
1.6569	12.0	48	0.9085	0.2042	0.9092	0.9535
1.384	12.5	50	0.8219	0.3838	0.8223	0.9068
1.384	13.0	52	1.3913	0.2558	1.3918	1.1798
1.384	13.5	54	1.2815	0.2852	1.2820	1.1323
1.384	14.0	56	0.7346	0.4095	0.7349	0.8572
1.384	14.5	58	0.6847	0.4148	0.6850	0.8276
1.0521	15.0	60	0.7324	0.4182	0.7326	0.8559
1.0521	15.5	62	0.9301	0.3275	0.9304	0.9646
1.0521	16.0	64	0.7825	0.4208	0.7828	0.8848
1.0521	16.5	66	0.6660	0.5040	0.6663	0.8163
1.0521	17.0	68	0.7226	0.4766	0.7229	0.8502
0.5077	17.5	70	0.7386	0.4966	0.7387	0.8595
0.5077	18.0	72	0.6899	0.5230	0.6899	0.8306
0.5077	18.5	74	0.7018	0.5207	0.7017	0.8377
0.5077	19.0	76	0.9034	0.4526	0.9036	0.9506
0.5077	19.5	78	0.9620	0.4174	0.9623	0.9809
0.347	20.0	80	0.7991	0.4970	0.7993	0.8941
0.347	20.5	82	0.7782	0.4837	0.7784	0.8822
0.347	21.0	84	0.8148	0.4597	0.8149	0.9027
0.347	21.5	86	0.7619	0.4784	0.7619	0.8729
0.347	22.0	88	0.8145	0.4654	0.8144	0.9024
0.2213	22.5	90	0.8827	0.4370	0.8826	0.9395
0.2213	23.0	92	0.7723	0.4767	0.7721	0.8787

Framework versions

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