Version23NewTestASAP_FineTuningBERT_AugV23_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: 0.5269
• Qwk: 0.5585
• Mse: 0.5260
• Rmse: 0.7253

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: 150

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	1.0	2	8.3983	0.0	8.3957	2.8975
No log	2.0	4	7.7029	0.0	7.7005	2.7750
No log	3.0	6	6.4804	0.0	6.4780	2.5452
No log	4.0	8	5.2430	0.0271	5.2408	2.2893
No log	5.0	10	4.0727	0.0	4.0708	2.0176
No log	6.0	12	3.1750	0.0	3.1732	1.7813
No log	7.0	14	2.5239	0.0	2.5221	1.5881
No log	8.0	16	1.9365	0.0365	1.9348	1.3910
No log	9.0	18	1.4916	0.0106	1.4901	1.2207
No log	10.0	20	1.2175	0.0106	1.2161	1.1028
No log	11.0	22	0.9932	0.0106	0.9918	0.9959
No log	12.0	24	0.8456	0.3134	0.8443	0.9189
No log	13.0	26	0.7295	0.4125	0.7284	0.8534
No log	14.0	28	0.6442	0.4473	0.6431	0.8019
No log	15.0	30	0.6141	0.5203	0.6130	0.7829
No log	16.0	32	0.5959	0.4648	0.5949	0.7713
No log	17.0	34	0.6381	0.5050	0.6368	0.7980
No log	18.0	36	0.5184	0.5372	0.5171	0.7191
No log	19.0	38	0.5895	0.5006	0.5884	0.7670
No log	20.0	40	0.5284	0.5425	0.5270	0.7259
No log	21.0	42	0.7067	0.4693	0.7056	0.8400
No log	22.0	44	0.5653	0.5199	0.5640	0.7510
No log	23.0	46	0.6061	0.5853	0.6049	0.7777
No log	24.0	48	0.5510	0.5842	0.5499	0.7415
No log	25.0	50	1.0922	0.4319	1.0910	1.0445
No log	26.0	52	0.5903	0.5933	0.5893	0.7676
No log	27.0	54	0.5942	0.6030	0.5932	0.7702
No log	28.0	56	0.8102	0.5393	0.8091	0.8995
No log	29.0	58	0.6498	0.5665	0.6490	0.8056
No log	30.0	60	0.6124	0.5733	0.6115	0.7820
No log	31.0	62	0.7376	0.5142	0.7365	0.8582
No log	32.0	64	0.6127	0.5783	0.6119	0.7822
No log	33.0	66	0.5723	0.6111	0.5714	0.7559
No log	34.0	68	0.5408	0.6368	0.5400	0.7348
No log	35.0	70	0.6060	0.6057	0.6051	0.7779
No log	36.0	72	0.6293	0.6081	0.6288	0.7930
No log	37.0	74	0.5774	0.6284	0.5768	0.7595
No log	38.0	76	0.5804	0.5931	0.5796	0.7613
No log	39.0	78	0.5765	0.5978	0.5759	0.7589
No log	40.0	80	1.2694	0.4347	1.2691	1.1265
No log	41.0	82	0.8455	0.5134	0.8450	0.9192
No log	42.0	84	0.6682	0.5235	0.6674	0.8170
No log	43.0	86	0.5972	0.5565	0.5964	0.7723
No log	44.0	88	0.7299	0.5559	0.7294	0.8541
No log	45.0	90	0.5544	0.6037	0.5537	0.7441
No log	46.0	92	0.5444	0.6064	0.5438	0.7374
No log	47.0	94	0.5402	0.6127	0.5397	0.7346
No log	48.0	96	0.5412	0.6072	0.5407	0.7353
No log	49.0	98	0.5417	0.5997	0.5412	0.7357
No log	50.0	100	0.5529	0.5788	0.5524	0.7432
No log	51.0	102	0.6530	0.5520	0.6526	0.8078
No log	52.0	104	0.6183	0.5509	0.6179	0.7860
No log	53.0	106	0.5756	0.5741	0.5749	0.7582
No log	54.0	108	0.7059	0.5453	0.7055	0.8399
No log	55.0	110	0.6391	0.5684	0.6386	0.7991
No log	56.0	112	0.5420	0.5906	0.5411	0.7356
No log	57.0	114	0.5423	0.5988	0.5415	0.7359
No log	58.0	116	0.6786	0.5326	0.6781	0.8235
No log	59.0	118	0.5748	0.5690	0.5740	0.7576
No log	60.0	120	0.5757	0.5539	0.5748	0.7582
No log	61.0	122	0.8121	0.5062	0.8117	0.9009
No log	62.0	124	0.8523	0.5031	0.8520	0.9230
No log	63.0	126	0.5380	0.5545	0.5372	0.7330
No log	64.0	128	0.5269	0.5585	0.5260	0.7253

Framework versions

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