Version3ASAP_FineTuningBERT_AugV12_k5_task1_organization_k5_k5_fold4

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: 0.8549
• Qwk: 0.6178
• Mse: 0.8549
• Rmse: 0.9246

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	11.0708	0.0166	11.0708	3.3273
No log	2.0	4	9.7417	0.0021	9.7417	3.1212
No log	3.0	6	8.4668	0.0018	8.4668	2.9098
No log	4.0	8	7.4044	0.0018	7.4044	2.7211
No log	5.0	10	6.3359	0.0018	6.3359	2.5171
No log	6.0	12	5.3670	0.0445	5.3670	2.3167
No log	7.0	14	4.5386	0.0156	4.5386	2.1304
No log	8.0	16	3.8757	0.0079	3.8757	1.9687
No log	9.0	18	3.2679	0.0079	3.2679	1.8077
No log	10.0	20	2.7570	0.0040	2.7570	1.6604
No log	11.0	22	2.3487	0.1179	2.3487	1.5326
No log	12.0	24	1.9222	0.0420	1.9222	1.3864
No log	13.0	26	1.6068	0.0420	1.6068	1.2676
No log	14.0	28	1.3849	0.0316	1.3849	1.1768
No log	15.0	30	1.1101	0.0212	1.1101	1.0536
No log	16.0	32	0.9479	0.0316	0.9479	0.9736
No log	17.0	34	0.8513	0.2601	0.8513	0.9226
No log	18.0	36	0.7868	0.2129	0.7868	0.8870
No log	19.0	38	0.6786	0.3549	0.6786	0.8238
No log	20.0	40	0.6848	0.3035	0.6848	0.8275
No log	21.0	42	0.7965	0.2382	0.7965	0.8925
No log	22.0	44	0.6465	0.3852	0.6465	0.8040
No log	23.0	46	0.7439	0.4860	0.7439	0.8625
No log	24.0	48	0.8200	0.4854	0.8200	0.9056
No log	25.0	50	0.6675	0.5974	0.6675	0.8170
No log	26.0	52	0.6893	0.6042	0.6893	0.8302
No log	27.0	54	0.9911	0.5676	0.9911	0.9955
No log	28.0	56	0.6598	0.6383	0.6598	0.8123
No log	29.0	58	0.6187	0.6630	0.6187	0.7866
No log	30.0	60	1.1921	0.5258	1.1921	1.0918
No log	31.0	62	0.5893	0.6772	0.5893	0.7676
No log	32.0	64	0.5806	0.6741	0.5806	0.7620
No log	33.0	66	0.7501	0.6221	0.7501	0.8661
No log	34.0	68	0.6106	0.6467	0.6106	0.7814
No log	35.0	70	1.3755	0.4868	1.3755	1.1728
No log	36.0	72	1.2697	0.5067	1.2697	1.1268
No log	37.0	74	0.6597	0.6431	0.6597	0.8122
No log	38.0	76	1.1212	0.5554	1.1212	1.0589
No log	39.0	78	1.4375	0.5064	1.4375	1.1990
No log	40.0	80	0.6695	0.6444	0.6695	0.8182
No log	41.0	82	0.6508	0.6576	0.6508	0.8067
No log	42.0	84	0.9245	0.6102	0.9245	0.9615
No log	43.0	86	0.6585	0.6304	0.6585	0.8115
No log	44.0	88	0.6926	0.6365	0.6926	0.8322
No log	45.0	90	0.8620	0.6192	0.8620	0.9284
No log	46.0	92	0.7562	0.6279	0.7562	0.8696
No log	47.0	94	1.1786	0.5620	1.1786	1.0856
No log	48.0	96	1.1252	0.5623	1.1252	1.0607
No log	49.0	98	0.6571	0.6388	0.6571	0.8106
No log	50.0	100	0.6293	0.6533	0.6293	0.7933
No log	51.0	102	0.8549	0.6178	0.8549	0.9246

Framework versions

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