Version22NewTestASAP_FineTuningBERT_AugV22_k5_task1_organization_k5_k5_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: 0.8897
• Qwk: 0.4626
• Mse: 0.8894
• Rmse: 0.9431

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	3	9.1249	0.0	9.1249	3.0207
No log	2.0	6	6.4571	0.0	6.4575	2.5412
No log	3.0	9	4.5661	0.0039	4.5666	2.1370
No log	4.0	12	3.2668	0.0	3.2673	1.8076
No log	5.0	15	2.4546	0.0360	2.4550	1.5668
No log	6.0	18	1.8095	0.0296	1.8100	1.3454
No log	7.0	21	1.4514	0.0107	1.4519	1.2049
No log	8.0	24	1.5773	0.0906	1.5778	1.2561
No log	9.0	27	1.1706	0.0475	1.1710	1.0821
No log	10.0	30	1.4654	0.1182	1.4658	1.2107
No log	11.0	33	1.0020	0.1229	1.0023	1.0011
No log	12.0	36	0.9916	0.2343	0.9919	0.9959
No log	13.0	39	0.7031	0.4152	0.7033	0.8386
No log	14.0	42	1.1188	0.3096	1.1190	1.0578
No log	15.0	45	0.6470	0.4413	0.6472	0.8045
No log	16.0	48	0.6823	0.4069	0.6825	0.8262
No log	17.0	51	0.6444	0.4508	0.6446	0.8028
No log	18.0	54	0.6332	0.4099	0.6334	0.7959
No log	19.0	57	0.6138	0.4771	0.6141	0.7836
No log	20.0	60	0.6682	0.3679	0.6685	0.8176
No log	21.0	63	0.6052	0.4771	0.6055	0.7781
No log	22.0	66	0.6287	0.4843	0.6289	0.7931
No log	23.0	69	0.6367	0.4797	0.6369	0.7981
No log	24.0	72	0.6546	0.4617	0.6548	0.8092
No log	25.0	75	0.6527	0.4348	0.6529	0.8080
No log	26.0	78	0.6735	0.4879	0.6737	0.8208
No log	27.0	81	0.6695	0.4886	0.6697	0.8184
No log	28.0	84	0.6197	0.5302	0.6199	0.7873
No log	29.0	87	0.6703	0.5152	0.6704	0.8188
No log	30.0	90	0.6822	0.5332	0.6823	0.8260
No log	31.0	93	0.7634	0.4803	0.7634	0.8737
No log	32.0	96	0.6853	0.5095	0.6854	0.8279
No log	33.0	99	0.6374	0.5317	0.6375	0.7984
No log	34.0	102	0.7943	0.4550	0.7943	0.8912
No log	35.0	105	0.6468	0.5344	0.6469	0.8043
No log	36.0	108	0.6371	0.5372	0.6372	0.7982
No log	37.0	111	0.6319	0.5527	0.6320	0.7950
No log	38.0	114	0.6884	0.5219	0.6884	0.8297
No log	39.0	117	0.7159	0.5068	0.7159	0.8461
No log	40.0	120	0.7026	0.4923	0.7026	0.8382
No log	41.0	123	0.7569	0.5034	0.7567	0.8699
No log	42.0	126	0.8155	0.4752	0.8152	0.9029
No log	43.0	129	0.6914	0.5233	0.6912	0.8314
No log	44.0	132	0.8076	0.4525	0.8074	0.8985
No log	45.0	135	0.6673	0.4971	0.6672	0.8168
No log	46.0	138	0.6988	0.5291	0.6986	0.8358
No log	47.0	141	0.8245	0.4935	0.8242	0.9079
No log	48.0	144	0.7806	0.5132	0.7804	0.8834
No log	49.0	147	0.8122	0.4991	0.8119	0.9010
No log	50.0	150	0.7885	0.5196	0.7881	0.8878
No log	51.0	153	0.9458	0.4912	0.9453	0.9723
No log	52.0	156	0.9098	0.4885	0.9094	0.9536
No log	53.0	159	0.8816	0.4783	0.8813	0.9388
No log	54.0	162	0.8011	0.5059	0.8009	0.8949
No log	55.0	165	0.8841	0.4753	0.8840	0.9402
No log	56.0	168	0.7255	0.4881	0.7254	0.8517
No log	57.0	171	0.8043	0.4985	0.8041	0.8967
No log	58.0	174	0.8381	0.4927	0.8378	0.9153
No log	59.0	177	0.8368	0.5069	0.8365	0.9146
No log	60.0	180	0.9250	0.4750	0.9246	0.9616
No log	61.0	183	0.8174	0.5030	0.8170	0.9039
No log	62.0	186	0.8865	0.5011	0.8861	0.9414
No log	63.0	189	0.8292	0.5256	0.8288	0.9104
No log	64.0	192	0.8713	0.5077	0.8709	0.9332
No log	65.0	195	0.8620	0.5051	0.8616	0.9282
No log	66.0	198	0.7571	0.5205	0.7568	0.8699
No log	67.0	201	0.8897	0.4626	0.8894	0.9431

Framework versions

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