Version19ASAP_FineTuningBERT_AugV19_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: 0.9222
• Qwk: 0.5375
• Mse: 0.9212
• Rmse: 0.9598

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	2	8.6091	0.0	8.6096	2.9342
No log	2.0	4	5.9640	0.0236	5.9643	2.4422
No log	3.0	6	4.6195	0.0045	4.6199	2.1494
No log	4.0	8	3.5608	0.0	3.5612	1.8871
No log	5.0	10	2.8663	0.0	2.8666	1.6931
No log	6.0	12	2.3110	0.1005	2.3114	1.5203
No log	7.0	14	1.7755	0.0107	1.7759	1.3326
No log	8.0	16	1.4128	0.0	1.4133	1.1888
No log	9.0	18	1.1792	0.0	1.1796	1.0861
No log	10.0	20	0.9868	0.0	0.9872	0.9936
No log	11.0	22	0.8551	0.3396	0.8555	0.9249
No log	12.0	24	0.8150	0.1043	0.8155	0.9030
No log	13.0	26	0.7615	0.0925	0.7619	0.8729
No log	14.0	28	0.7520	0.0925	0.7523	0.8673
No log	15.0	30	0.8481	0.0883	0.8482	0.9210
No log	16.0	32	0.6730	0.2633	0.6728	0.8203
No log	17.0	34	0.6372	0.5541	0.6367	0.7979
No log	18.0	36	0.6751	0.5552	0.6744	0.8212
No log	19.0	38	0.6994	0.5454	0.6985	0.8358
No log	20.0	40	0.8174	0.5028	0.8164	0.9036
No log	21.0	42	0.9374	0.4832	0.9362	0.9676
No log	22.0	44	1.1207	0.4508	1.1195	1.0581
No log	23.0	46	0.9431	0.5110	0.9419	0.9705
No log	24.0	48	0.8850	0.5594	0.8838	0.9401
No log	25.0	50	0.9857	0.5212	0.9844	0.9922
No log	26.0	52	0.6241	0.6278	0.6233	0.7895
No log	27.0	54	0.8711	0.5419	0.8702	0.9329
No log	28.0	56	1.4097	0.4349	1.4086	1.1868
No log	29.0	58	0.6073	0.6119	0.6068	0.7790
No log	30.0	60	0.5749	0.6241	0.5744	0.7579
No log	31.0	62	1.3681	0.4796	1.3671	1.1692
No log	32.0	64	1.4350	0.4706	1.4338	1.1974
No log	33.0	66	0.6212	0.6267	0.6206	0.7878
No log	34.0	68	0.5639	0.6561	0.5634	0.7506
No log	35.0	70	0.8115	0.5919	0.8106	0.9004
No log	36.0	72	0.8938	0.5613	0.8929	0.9449
No log	37.0	74	0.6572	0.6054	0.6565	0.8103
No log	38.0	76	0.9720	0.5242	0.9710	0.9854
No log	39.0	78	0.9011	0.5476	0.9001	0.9487
No log	40.0	80	0.6509	0.6110	0.6502	0.8063
No log	41.0	82	0.6971	0.6145	0.6962	0.8344
No log	42.0	84	0.9584	0.5490	0.9573	0.9784
No log	43.0	86	0.9103	0.5536	0.9093	0.9536
No log	44.0	88	0.6691	0.6161	0.6684	0.8176
No log	45.0	90	0.7851	0.5720	0.7843	0.8856
No log	46.0	92	1.0644	0.5125	1.0634	1.0312
No log	47.0	94	0.8422	0.5673	0.8413	0.9172
No log	48.0	96	0.6230	0.6313	0.6223	0.7889
No log	49.0	98	0.7121	0.5993	0.7114	0.8434
No log	50.0	100	1.0733	0.5121	1.0723	1.0355
No log	51.0	102	0.9953	0.5204	0.9944	0.9972
No log	52.0	104	0.7121	0.5925	0.7114	0.8434
No log	53.0	106	0.7495	0.5808	0.7487	0.8653
No log	54.0	108	0.9222	0.5375	0.9212	0.9598

Framework versions

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