ASAP_FineTuningBERT_AugV14_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: 0.8497
• Qwk: 0.3966
• Mse: 0.8485
• Rmse: 0.9211

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	5	8.1321	0.0	8.1295	2.8512
No log	2.0	10	4.5392	0.0	4.5370	2.1300
No log	3.0	15	2.4253	0.0008	2.4235	1.5567
No log	4.0	20	1.3315	0.0	1.3300	1.1532
No log	5.0	25	1.0503	0.0	1.0490	1.0242
No log	6.0	30	1.0201	0.0	1.0188	1.0094
No log	7.0	35	1.3639	0.1089	1.3623	1.1672
No log	8.0	40	0.7009	0.3505	0.7000	0.8367
No log	9.0	45	0.6217	0.4536	0.6210	0.7880
No log	10.0	50	0.6210	0.4812	0.6203	0.7876
No log	11.0	55	0.7873	0.3590	0.7861	0.8866
No log	12.0	60	1.0020	0.3433	1.0009	1.0005
No log	13.0	65	0.7621	0.4468	0.7610	0.8724
No log	14.0	70	0.8290	0.4211	0.8280	0.9099
No log	15.0	75	0.9144	0.3387	0.9132	0.9556
No log	16.0	80	0.8671	0.4052	0.8665	0.9308
No log	17.0	85	0.8755	0.3927	0.8746	0.9352
No log	18.0	90	0.8636	0.4164	0.8624	0.9286
No log	19.0	95	0.8986	0.3277	0.8973	0.9473
No log	20.0	100	1.0249	0.3529	1.0234	1.0116
No log	21.0	105	0.9556	0.3586	0.9542	0.9769
No log	22.0	110	0.8603	0.4105	0.8591	0.9269
No log	23.0	115	0.8487	0.4383	0.8477	0.9207
No log	24.0	120	0.8726	0.3984	0.8717	0.9336
No log	25.0	125	0.8232	0.4131	0.8221	0.9067
No log	26.0	130	0.9169	0.3900	0.9161	0.9572
No log	27.0	135	1.0698	0.3536	1.0683	1.0336
No log	28.0	140	0.7608	0.4442	0.7599	0.8717
No log	29.0	145	0.7792	0.4220	0.7782	0.8821
No log	30.0	150	0.8497	0.3966	0.8485	0.9211

Framework versions

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