Version_weird_ASAP_FineTuningBERT_AugV12_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: 0.7854
• Qwk: 0.6119
• Mse: 0.7849
• Rmse: 0.8860

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	9.7366	0.0019	9.7341	3.1200
No log	2.0	4	6.7742	0.0	6.7717	2.6022
No log	3.0	6	4.9301	0.0215	4.9278	2.2199
No log	4.0	8	3.6455	0.0	3.6438	1.9089
No log	5.0	10	2.9507	0.0	2.9488	1.7172
No log	6.0	12	2.2522	0.1368	2.2504	1.5001
No log	7.0	14	1.6882	0.0211	1.6866	1.2987
No log	8.0	16	1.2445	0.0	1.2430	1.1149
No log	9.0	18	1.1161	0.0	1.1147	1.0558
No log	10.0	20	0.9311	0.2802	0.9298	0.9643
No log	11.0	22	0.8784	0.0873	0.8772	0.9366
No log	12.0	24	0.9432	0.0143	0.9422	0.9707
No log	13.0	26	1.0114	0.0	1.0103	1.0051
No log	14.0	28	1.1169	0.0	1.1160	1.0564
No log	15.0	30	1.1922	0.0143	1.1914	1.0915
No log	16.0	32	1.3580	0.2887	1.3573	1.1650
No log	17.0	34	1.1579	0.3388	1.1572	1.0757
No log	18.0	36	0.8333	0.4221	0.8326	0.9125
No log	19.0	38	1.0902	0.3982	1.0897	1.0439
No log	20.0	40	0.6886	0.5381	0.6880	0.8295
No log	21.0	42	0.4147	0.6672	0.4140	0.6434
No log	22.0	44	0.8871	0.5626	0.8867	0.9417
No log	23.0	46	0.8801	0.6008	0.8798	0.9379
No log	24.0	48	0.4773	0.7254	0.4769	0.6906
No log	25.0	50	0.4673	0.7061	0.4668	0.6833
No log	26.0	52	0.8353	0.6218	0.8350	0.9138
No log	27.0	54	0.6241	0.6775	0.6237	0.7898
No log	28.0	56	0.3975	0.7450	0.3969	0.6300
No log	29.0	58	0.6754	0.6502	0.6750	0.8216
No log	30.0	60	2.0530	0.3927	2.0527	1.4327
No log	31.0	62	1.5378	0.4564	1.5375	1.2400
No log	32.0	64	0.4413	0.6977	0.4407	0.6639
No log	33.0	66	0.3897	0.7273	0.3891	0.6238
No log	34.0	68	0.6623	0.6305	0.6618	0.8135
No log	35.0	70	1.1973	0.5186	1.1969	1.0940
No log	36.0	72	0.7535	0.6139	0.7530	0.8678
No log	37.0	74	0.4153	0.7098	0.4147	0.6440
No log	38.0	76	0.5182	0.6681	0.5177	0.7195
No log	39.0	78	1.2103	0.5328	1.2099	1.1000
No log	40.0	80	1.3066	0.5288	1.3062	1.1429
No log	41.0	82	0.6456	0.6632	0.6451	0.8032
No log	42.0	84	0.3997	0.7467	0.3991	0.6318
No log	43.0	86	0.4312	0.7078	0.4307	0.6562
No log	44.0	88	0.9140	0.5813	0.9135	0.9558
No log	45.0	90	1.2731	0.5050	1.2726	1.1281
No log	46.0	92	0.8951	0.5898	0.8946	0.9458
No log	47.0	94	0.5963	0.6754	0.5958	0.7719
No log	48.0	96	0.7580	0.6416	0.7575	0.8704
No log	49.0	98	1.1626	0.5432	1.1621	1.0780
No log	50.0	100	0.9226	0.5884	0.9221	0.9603
No log	51.0	102	0.6106	0.6686	0.6101	0.7811
No log	52.0	104	0.6670	0.6480	0.6665	0.8164
No log	53.0	106	0.9506	0.5797	0.9501	0.9747
No log	54.0	108	0.8086	0.6048	0.8081	0.8990
No log	55.0	110	0.8097	0.5970	0.8092	0.8995
No log	56.0	112	0.9118	0.5715	0.9113	0.9546
No log	57.0	114	0.9276	0.5816	0.9270	0.9628
No log	58.0	116	0.8875	0.5976	0.8870	0.9418
No log	59.0	118	0.7571	0.6236	0.7566	0.8699
No log	60.0	120	0.9790	0.5715	0.9785	0.9892
No log	61.0	122	1.0024	0.5608	1.0019	1.0010
No log	62.0	124	0.7854	0.6119	0.7849	0.8860

Framework versions

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