Version3ASAP_FineTuningBERT_AugV12_k10_task1_organization_k10_k10_fold0

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: 0.6628
• Qwk: 0.6169
• Mse: 0.6628
• Rmse: 0.8141

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	8.6894	0.0	8.6894	2.9478
No log	2.0	6	7.1508	0.0	7.1508	2.6741
No log	3.0	9	6.1796	-0.0223	6.1796	2.4859
No log	4.0	12	5.0489	0.0077	5.0489	2.2470
No log	5.0	15	3.9470	0.0039	3.9470	1.9867
No log	6.0	18	2.9441	0.0	2.9441	1.7158
No log	7.0	21	2.1053	0.0601	2.1053	1.4510
No log	8.0	24	1.4740	0.0316	1.4740	1.2141
No log	9.0	27	1.1165	0.0106	1.1165	1.0566
No log	10.0	30	0.8320	0.2712	0.8320	0.9121
No log	11.0	33	0.7768	0.1122	0.7768	0.8813
No log	12.0	36	0.8030	0.0521	0.8030	0.8961
No log	13.0	39	0.8769	0.0521	0.8769	0.9364
No log	14.0	42	1.0193	0.2721	1.0193	1.0096
No log	15.0	45	1.0655	0.3087	1.0655	1.0322
No log	16.0	48	1.0367	0.3652	1.0367	1.0182
No log	17.0	51	0.5579	0.4677	0.5579	0.7470
No log	18.0	54	0.7125	0.5098	0.7125	0.8441
No log	19.0	57	0.6865	0.5073	0.6865	0.8285
No log	20.0	60	0.9630	0.5471	0.9630	0.9813
No log	21.0	63	0.9938	0.4577	0.9938	0.9969
No log	22.0	66	0.9722	0.4831	0.9722	0.9860
No log	23.0	69	0.6949	0.5642	0.6949	0.8336
No log	24.0	72	1.0135	0.5395	1.0135	1.0067
No log	25.0	75	0.7379	0.6330	0.7379	0.8590
No log	26.0	78	0.5702	0.6383	0.5702	0.7551
No log	27.0	81	0.9379	0.5543	0.9379	0.9685
No log	28.0	84	0.5647	0.6465	0.5647	0.7515
No log	29.0	87	0.8128	0.5990	0.8128	0.9016
No log	30.0	90	0.5949	0.6378	0.5949	0.7713
No log	31.0	93	0.7179	0.6205	0.7179	0.8473
No log	32.0	96	0.7752	0.6029	0.7752	0.8804
No log	33.0	99	0.5903	0.6362	0.5903	0.7683
No log	34.0	102	0.8385	0.5874	0.8385	0.9157
No log	35.0	105	0.9683	0.5422	0.9683	0.9840
No log	36.0	108	0.5062	0.6342	0.5062	0.7115
No log	37.0	111	0.6541	0.5965	0.6541	0.8088
No log	38.0	114	0.6101	0.6094	0.6101	0.7811
No log	39.0	117	0.5205	0.6267	0.5205	0.7214
No log	40.0	120	0.5243	0.6151	0.5243	0.7241
No log	41.0	123	0.7361	0.6150	0.7361	0.8579
No log	42.0	126	0.4942	0.6253	0.4942	0.7030
No log	43.0	129	0.6450	0.6060	0.6450	0.8031
No log	44.0	132	0.6055	0.6231	0.6055	0.7781
No log	45.0	135	0.5379	0.6308	0.5379	0.7334
No log	46.0	138	0.6156	0.6335	0.6156	0.7846
No log	47.0	141	0.5515	0.6290	0.5515	0.7426
No log	48.0	144	0.6628	0.6169	0.6628	0.8141

Framework versions

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