ASAP_FineTuningBERT_AugV5_k10_task1_organization_fold0

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: 0.5805
• Qwk: 0.6413
• Mse: 0.5805
• Rmse: 0.7619

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	2.0	2	8.9334	0.0	8.9334	2.9889
No log	4.0	4	7.7251	0.0	7.7251	2.7794
No log	6.0	6	6.9126	0.0	6.9126	2.6292
No log	8.0	8	6.0855	-0.0067	6.0855	2.4669
8.7869	10.0	10	5.2885	0.0115	5.2885	2.2997
8.7869	12.0	12	4.4876	0.0039	4.4876	2.1184
8.7869	14.0	14	3.8555	0.0	3.8555	1.9635
8.7869	16.0	16	3.1151	0.0	3.1151	1.7650
8.7869	18.0	18	2.4646	0.0902	2.4646	1.5699
4.8367	20.0	20	1.9258	0.0511	1.9258	1.3877
4.8367	22.0	22	1.5573	0.0316	1.5573	1.2479
4.8367	24.0	24	1.2277	0.0316	1.2277	1.1080
4.8367	26.0	26	1.0600	0.0484	1.0600	1.0296
4.8367	28.0	28	0.8888	0.0484	0.8888	0.9428
2.5363	30.0	30	0.7514	0.4218	0.7514	0.8668
2.5363	32.0	32	0.6534	0.4791	0.6534	0.8083
2.5363	34.0	34	0.7060	0.4858	0.7060	0.8402
2.5363	36.0	36	0.5431	0.5111	0.5431	0.7370
2.5363	38.0	38	0.5125	0.4976	0.5125	0.7159
1.2777	40.0	40	0.5039	0.5206	0.5039	0.7099
1.2777	42.0	42	0.5139	0.4642	0.5139	0.7169
1.2777	44.0	44	0.5084	0.5505	0.5084	0.7130
1.2777	46.0	46	0.5008	0.5513	0.5008	0.7077
1.2777	48.0	48	0.4937	0.6186	0.4937	0.7026
0.6754	50.0	50	0.4821	0.5879	0.4821	0.6943
0.6754	52.0	52	0.5527	0.5948	0.5527	0.7434
0.6754	54.0	54	0.5330	0.6053	0.5330	0.7301
0.6754	56.0	56	0.4867	0.6329	0.4867	0.6976
0.6754	58.0	58	0.5515	0.6071	0.5515	0.7426
0.3859	60.0	60	0.5178	0.6378	0.5178	0.7196
0.3859	62.0	62	0.5138	0.6373	0.5138	0.7168
0.3859	64.0	64	0.5630	0.6289	0.5630	0.7503
0.3859	66.0	66	0.5752	0.6292	0.5752	0.7584
0.3859	68.0	68	0.6784	0.5942	0.6784	0.8236
0.191	70.0	70	0.6061	0.6268	0.6061	0.7785
0.191	72.0	72	0.6469	0.6089	0.6469	0.8043
0.191	74.0	74	0.5509	0.6231	0.5509	0.7422
0.191	76.0	76	0.5431	0.6220	0.5431	0.7370
0.191	78.0	78	0.6168	0.6288	0.6168	0.7853
0.1171	80.0	80	0.5948	0.6323	0.5948	0.7712
0.1171	82.0	82	0.6731	0.6193	0.6731	0.8204
0.1171	84.0	84	0.6456	0.6204	0.6456	0.8035
0.1171	86.0	86	0.5760	0.6413	0.5760	0.7590
0.1171	88.0	88	0.5489	0.6397	0.5489	0.7408
0.1069	90.0	90	0.5869	0.6396	0.5869	0.7661
0.1069	92.0	92	0.6605	0.6047	0.6605	0.8127
0.1069	94.0	94	0.6535	0.6093	0.6535	0.8084
0.1069	96.0	96	0.6031	0.6271	0.6031	0.7766
0.1069	98.0	98	0.5848	0.6402	0.5848	0.7647
0.0854	100.0	100	0.5805	0.6413	0.5805	0.7619

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1