Version2ASAP_FineTuningBERT_AugV5_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: 0.8740
• Qwk: 0.4181
• Mse: 0.8741
• Rmse: 0.9350

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	8	6.1672	0.0004	6.1674	2.4834
No log	2.0	16	2.3276	0.1353	2.3280	1.5258
No log	3.0	24	1.2013	0.0	1.2018	1.0963
No log	4.0	32	1.6262	0.1539	1.6266	1.2754
No log	5.0	40	0.8388	0.3749	0.8391	0.9160
No log	6.0	48	0.8037	0.2634	0.8040	0.8967
No log	7.0	56	1.0770	0.3703	1.0772	1.0379
No log	8.0	64	0.6944	0.4195	0.6945	0.8334
No log	9.0	72	0.7164	0.4583	0.7165	0.8465
No log	10.0	80	0.8141	0.4819	0.8137	0.9021
No log	11.0	88	0.8503	0.4533	0.8501	0.9220
No log	12.0	96	0.8391	0.4753	0.8389	0.9159
No log	13.0	104	0.8459	0.4807	0.8458	0.9197
No log	14.0	112	0.9768	0.4332	0.9772	0.9886
No log	15.0	120	0.8962	0.4442	0.8966	0.9469
No log	16.0	128	0.8922	0.4375	0.8923	0.9446
No log	17.0	136	0.8390	0.4288	0.8391	0.9160
No log	18.0	144	0.8757	0.3857	0.8760	0.9359
No log	19.0	152	0.8252	0.4364	0.8252	0.9084
No log	20.0	160	0.8085	0.4527	0.8085	0.8992
No log	21.0	168	0.9471	0.4345	0.9471	0.9732
No log	22.0	176	0.8389	0.4025	0.8391	0.9160
No log	23.0	184	0.8059	0.3923	0.8061	0.8978
No log	24.0	192	0.8653	0.4438	0.8654	0.9303
No log	25.0	200	0.7807	0.4413	0.7807	0.8836
No log	26.0	208	0.8250	0.4258	0.8252	0.9084
No log	27.0	216	0.7748	0.4533	0.7751	0.8804
No log	28.0	224	0.7487	0.4532	0.7489	0.8654
No log	29.0	232	0.8374	0.4073	0.8376	0.9152
No log	30.0	240	0.8740	0.4181	0.8741	0.9350

Framework versions

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