Version19ASAP_FineTuningBERT_AugV19_k3_task1_organization_k3_k3_fold1

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: 1.0376
• Qwk: 0.5471
• Mse: 1.0375
• Rmse: 1.0186

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	11.8886	-0.0059	11.8858	3.4476
No log	2.0	2	10.5573	-0.0052	10.5546	3.2488
No log	3.0	3	9.1684	0.0	9.1658	3.0275
No log	4.0	4	8.1056	0.0	8.1032	2.8466
No log	5.0	5	7.1750	0.0	7.1726	2.6782
No log	6.0	6	6.4704	0.0	6.4681	2.5432
No log	7.0	7	5.8063	0.0017	5.8039	2.4091
No log	8.0	8	5.0772	0.0160	5.0750	2.2528
No log	9.0	9	4.4349	0.0118	4.4328	2.1054
No log	10.0	10	3.9162	0.0	3.9142	1.9784
No log	11.0	11	3.5697	0.0	3.5677	1.8888
No log	12.0	12	3.1346	0.0	3.1327	1.7699
No log	13.0	13	2.6957	0.0	2.6940	1.6413
No log	14.0	14	2.3972	0.0215	2.3955	1.5477
No log	15.0	15	2.2043	0.0717	2.2026	1.4841
No log	16.0	16	1.9970	0.0583	1.9954	1.4126
No log	17.0	17	1.7335	0.0418	1.7319	1.3160
No log	18.0	18	1.5635	0.0315	1.5619	1.2498
No log	19.0	19	1.4379	0.0106	1.4364	1.1985
No log	20.0	20	1.3608	0.0	1.3594	1.1659
No log	21.0	21	1.2338	0.0	1.2324	1.1101
No log	22.0	22	1.1181	0.0	1.1167	1.0567
No log	23.0	23	1.0266	0.0	1.0252	1.0125
No log	24.0	24	0.9596	0.0	0.9583	0.9789
No log	25.0	25	0.9111	0.0065	0.9098	0.9539
No log	26.0	26	0.9186	0.1127	0.9173	0.9578
No log	27.0	27	0.8718	0.2581	0.8706	0.9331
No log	28.0	28	0.8018	0.3005	0.8006	0.8947
No log	29.0	29	0.7580	0.3101	0.7568	0.8700
No log	30.0	30	0.7352	0.2843	0.7340	0.8567
No log	31.0	31	0.7477	0.2412	0.7466	0.8640
No log	32.0	32	0.7385	0.2438	0.7374	0.8587
No log	33.0	33	0.6609	0.2995	0.6599	0.8123
No log	34.0	34	0.6383	0.3406	0.6373	0.7983
No log	35.0	35	0.6464	0.3315	0.6455	0.8034
No log	36.0	36	0.6550	0.3396	0.6541	0.8088
No log	37.0	37	0.6008	0.3754	0.6000	0.7746
No log	38.0	38	0.5103	0.4665	0.5095	0.7138
No log	39.0	39	0.4875	0.4826	0.4867	0.6977
No log	40.0	40	0.5258	0.4763	0.5251	0.7246
No log	41.0	41	0.5115	0.5250	0.5108	0.7147
No log	42.0	42	0.4528	0.6609	0.4522	0.6724
No log	43.0	43	0.5005	0.6751	0.4999	0.7071
No log	44.0	44	0.6292	0.6060	0.6286	0.7929
No log	45.0	45	0.6268	0.6331	0.6262	0.7913
No log	46.0	46	0.5131	0.6793	0.5126	0.7160
No log	47.0	47	0.5623	0.6793	0.5619	0.7496
No log	48.0	48	0.7609	0.5709	0.7604	0.8720
No log	49.0	49	0.7692	0.5780	0.7687	0.8768
No log	50.0	50	0.6833	0.6413	0.6829	0.8264
No log	51.0	51	0.6326	0.6613	0.6322	0.7951
No log	52.0	52	0.7688	0.6086	0.7684	0.8766
No log	53.0	53	0.8941	0.5637	0.8938	0.9454
No log	54.0	54	0.8838	0.5684	0.8835	0.9399
No log	55.0	55	0.7357	0.6385	0.7355	0.8576
No log	56.0	56	0.7961	0.6221	0.7959	0.8921
No log	57.0	57	1.0451	0.5591	1.0448	1.0222
No log	58.0	58	1.1874	0.5036	1.1871	1.0895
No log	59.0	59	1.0587	0.5716	1.0585	1.0288
No log	60.0	60	0.7840	0.6624	0.7839	0.8854
No log	61.0	61	0.7670	0.6690	0.7670	0.8758
No log	62.0	62	0.9499	0.6160	0.9499	0.9746
No log	63.0	63	1.1110	0.5695	1.1109	1.0540
No log	64.0	64	1.0894	0.5716	1.0893	1.0437
No log	65.0	65	0.8785	0.6071	0.8784	0.9373
No log	66.0	66	0.8667	0.5973	0.8665	0.9309
No log	67.0	67	1.0376	0.5471	1.0375	1.0186

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.3.1
• Tokenizers 0.21.0