Version19ASAP_FineTuningBERT_AugV19_k5_task1_organization_k5_k5_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: 0.9819
• Qwk: 0.5596
• Mse: 0.9824
• Rmse: 0.9911

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	9.7538	0.0078	9.7509	3.1226
No log	2.0	4	8.8299	0.0	8.8273	2.9711
No log	3.0	6	7.9287	0.0	7.9262	2.8154
No log	4.0	8	6.8318	0.0	6.8294	2.6133
No log	5.0	10	5.6801	0.0561	5.6780	2.3828
No log	6.0	12	4.8420	0.0010	4.8399	2.2000
No log	7.0	14	3.7989	0.0	3.7969	1.9486
No log	8.0	16	3.0827	0.0	3.0809	1.7553
No log	9.0	18	2.5164	0.0246	2.5146	1.5858
No log	10.0	20	1.9810	0.0781	1.9795	1.4069
No log	11.0	22	1.5480	0.0198	1.5465	1.2436
No log	12.0	24	1.1875	0.0	1.1862	1.0891
No log	13.0	26	0.9927	0.0	0.9914	0.9957
No log	14.0	28	0.8801	0.1923	0.8789	0.9375
No log	15.0	30	0.8462	0.1434	0.8451	0.9193
No log	16.0	32	0.9353	0.0642	0.9344	0.9666
No log	17.0	34	1.0876	0.0286	1.0868	1.0425
No log	18.0	36	1.2291	0.0286	1.2284	1.1083
No log	19.0	38	1.5517	0.1425	1.5511	1.2454
No log	20.0	40	1.3075	0.2809	1.3071	1.1433
No log	21.0	42	1.3492	0.2716	1.3489	1.1614
No log	22.0	44	1.5978	0.2307	1.5977	1.2640
No log	23.0	46	1.1600	0.3753	1.1600	1.0770
No log	24.0	48	0.9392	0.4730	0.9391	0.9691
No log	25.0	50	0.7380	0.5313	0.7379	0.8590
No log	26.0	52	0.5436	0.6463	0.5431	0.7370
No log	27.0	54	0.9093	0.5821	0.9096	0.9537
No log	28.0	56	0.6954	0.6676	0.6955	0.8339
No log	29.0	58	1.0544	0.5595	1.0548	1.0270
No log	30.0	60	1.0671	0.5713	1.0676	1.0332
No log	31.0	62	0.9455	0.5800	0.9459	0.9726
No log	32.0	64	0.8636	0.5917	0.8640	0.9295
No log	33.0	66	1.1104	0.5532	1.1110	1.0541
No log	34.0	68	0.6503	0.6532	0.6504	0.8065
No log	35.0	70	1.1757	0.5397	1.1765	1.0847
No log	36.0	72	0.8650	0.5689	0.8654	0.9303
No log	37.0	74	1.1409	0.5253	1.1415	1.0684
No log	38.0	76	0.9644	0.5663	0.9648	0.9823
No log	39.0	78	1.4796	0.4466	1.4804	1.2167
No log	40.0	80	0.9063	0.5389	0.9066	0.9522
No log	41.0	82	0.9880	0.5383	0.9884	0.9942
No log	42.0	84	1.5138	0.4749	1.5147	1.2307
No log	43.0	86	0.6040	0.6342	0.6040	0.7772
No log	44.0	88	0.5981	0.6395	0.5981	0.7733
No log	45.0	90	1.0579	0.5439	1.0584	1.0288
No log	46.0	92	0.8071	0.6209	0.8073	0.8985
No log	47.0	94	0.8317	0.6059	0.8319	0.9121
No log	48.0	96	0.9819	0.5596	0.9824	0.9911

Framework versions

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