ASAP_FineTuningBERT_AugV9_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: 0.7829
• Qwk: 0.4190
• Mse: 0.7813
• Rmse: 0.8839

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	7.9758	0.0	7.9732	2.8237
No log	2.0	6	5.7535	0.0158	5.7514	2.3982
No log	3.0	9	3.9215	0.0	3.9194	1.9798
No log	4.0	12	3.1039	0.0025	3.1022	1.7613
No log	5.0	15	2.3251	0.1028	2.3232	1.5242
No log	6.0	18	1.6417	0.0645	1.6401	1.2806
No log	7.0	21	1.4713	0.0211	1.4697	1.2123
No log	8.0	24	1.3191	0.0	1.3175	1.1478
No log	9.0	27	1.0546	0.0106	1.0529	1.0261
No log	10.0	30	0.9564	0.1151	0.9547	0.9771
No log	11.0	33	0.9715	0.1981	0.9695	0.9846
No log	12.0	36	0.7925	0.3502	0.7907	0.8892
No log	13.0	39	1.5808	0.1685	1.5780	1.2562
No log	14.0	42	0.8948	0.3512	0.8926	0.9448
No log	15.0	45	1.0834	0.3027	1.0810	1.0397
No log	16.0	48	0.7796	0.3778	0.7776	0.8818
No log	17.0	51	1.0215	0.3077	1.0192	1.0096
No log	18.0	54	0.8540	0.3390	0.8520	0.9230
No log	19.0	57	0.8051	0.3633	0.8031	0.8962
No log	20.0	60	0.7496	0.4186	0.7476	0.8647
No log	21.0	63	1.0352	0.3665	1.0327	1.0162
No log	22.0	66	0.8155	0.3904	0.8132	0.9018
No log	23.0	69	1.3717	0.2322	1.3692	1.1701
No log	24.0	72	0.8400	0.3300	0.8379	0.9154
No log	25.0	75	0.8448	0.3176	0.8427	0.9180
No log	26.0	78	1.0377	0.2928	1.0354	1.0175
No log	27.0	81	0.8015	0.3337	0.7995	0.8942
No log	28.0	84	0.9095	0.3579	0.9074	0.9526
No log	29.0	87	0.7741	0.3623	0.7723	0.8788
No log	30.0	90	0.7772	0.3834	0.7754	0.8805
No log	31.0	93	0.9983	0.3380	0.9962	0.9981
No log	32.0	96	0.7701	0.3621	0.7684	0.8766
No log	33.0	99	0.7239	0.3796	0.7224	0.8499
No log	34.0	102	0.9611	0.3359	0.9592	0.9794
No log	35.0	105	0.7100	0.4685	0.7086	0.8418
No log	36.0	108	0.7681	0.4266	0.7666	0.8755
No log	37.0	111	0.7117	0.4426	0.7102	0.8427
No log	38.0	114	0.7591	0.4122	0.7575	0.8704
No log	39.0	117	0.7193	0.3791	0.7178	0.8472
No log	40.0	120	0.7360	0.3578	0.7344	0.8570
No log	41.0	123	0.7187	0.3837	0.7172	0.8469
No log	42.0	126	0.7652	0.3724	0.7636	0.8738
No log	43.0	129	0.7134	0.4102	0.7118	0.8437
No log	44.0	132	0.8811	0.3746	0.8792	0.9376
No log	45.0	135	0.7104	0.4480	0.7087	0.8419
No log	46.0	138	0.8640	0.3878	0.8622	0.9285
No log	47.0	141	0.7629	0.4224	0.7613	0.8725
No log	48.0	144	0.6632	0.4122	0.6619	0.8136
No log	49.0	147	0.7407	0.4135	0.7393	0.8598
No log	50.0	150	0.6963	0.4121	0.6949	0.8336
No log	51.0	153	0.7146	0.3904	0.7132	0.8445
No log	52.0	156	0.7392	0.4021	0.7377	0.8589
No log	53.0	159	0.7536	0.4183	0.7520	0.8672
No log	54.0	162	0.6897	0.4259	0.6883	0.8296
No log	55.0	165	0.7829	0.4190	0.7813	0.8839

Framework versions

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