Version3ASAP_FineTuningBERT_AugV14_k3_task1_organization_k3_k3_fold2

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: 0.8353
• Qwk: 0.3751
• Mse: 0.8351
• Rmse: 0.9138

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	3	10.6206	0.0	10.6206	3.2589
No log	2.0	6	8.2873	0.0	8.2875	2.8788
No log	3.0	9	5.4075	0.0427	5.4080	2.3255
No log	4.0	12	3.2529	0.0	3.2533	1.8037
No log	5.0	15	2.2975	0.1340	2.2980	1.5159
No log	6.0	18	1.5384	0.0280	1.5389	1.2405
No log	7.0	21	0.9935	0.0280	0.9939	0.9969
No log	8.0	24	1.7822	0.1538	1.7827	1.3352
No log	9.0	27	0.9574	0.1332	0.9578	0.9786
No log	10.0	30	0.8269	0.2465	0.8271	0.9095
No log	11.0	33	1.4600	0.2002	1.4603	1.2084
No log	12.0	36	0.9397	0.2987	0.9398	0.9694
No log	13.0	39	0.7336	0.3786	0.7336	0.8565
No log	14.0	42	0.7625	0.2827	0.7625	0.8732
No log	15.0	45	0.7398	0.2814	0.7398	0.8601
No log	16.0	48	0.7304	0.3618	0.7302	0.8545
No log	17.0	51	0.7276	0.4132	0.7274	0.8529
No log	18.0	54	0.8312	0.3803	0.8310	0.9116
No log	19.0	57	0.9161	0.3789	0.9156	0.9569
No log	20.0	60	0.9202	0.3931	0.9198	0.9591
No log	21.0	63	0.8285	0.3928	0.8281	0.9100
No log	22.0	66	0.8544	0.3787	0.8539	0.9241
No log	23.0	69	1.0998	0.3261	1.0994	1.0485
No log	24.0	72	0.9134	0.4227	0.9128	0.9554
No log	25.0	75	1.0164	0.3631	1.0159	1.0079
No log	26.0	78	0.8835	0.4073	0.8830	0.9397
No log	27.0	81	0.9027	0.3731	0.9023	0.9499
No log	28.0	84	0.9577	0.3480	0.9573	0.9784
No log	29.0	87	0.9042	0.3597	0.9038	0.9507
No log	30.0	90	0.9734	0.3722	0.9731	0.9865
No log	31.0	93	0.8613	0.4075	0.8611	0.9279
No log	32.0	96	1.0465	0.3690	1.0463	1.0229
No log	33.0	99	0.8563	0.4142	0.8561	0.9252
No log	34.0	102	0.9754	0.3586	0.9752	0.9875
No log	35.0	105	0.8597	0.3887	0.8595	0.9271
No log	36.0	108	0.9437	0.3495	0.9435	0.9713
No log	37.0	111	0.8340	0.4019	0.8338	0.9131
No log	38.0	114	0.9467	0.3468	0.9466	0.9729
No log	39.0	117	0.8395	0.3804	0.8393	0.9162
No log	40.0	120	0.8911	0.3913	0.8910	0.9439
No log	41.0	123	0.8413	0.4169	0.8412	0.9172
No log	42.0	126	0.8922	0.4061	0.8922	0.9446
No log	43.0	129	0.7965	0.4647	0.7964	0.8924
No log	44.0	132	0.9629	0.3866	0.9627	0.9812
No log	45.0	135	0.8170	0.4503	0.8168	0.9038
No log	46.0	138	0.7512	0.4221	0.7509	0.8666
No log	47.0	141	0.8270	0.4394	0.8268	0.9093
No log	48.0	144	0.7959	0.4516	0.7957	0.8920
No log	49.0	147	0.9125	0.3951	0.9124	0.9552
No log	50.0	150	0.7549	0.4639	0.7547	0.8687
No log	51.0	153	0.8440	0.4269	0.8439	0.9186
No log	52.0	156	0.8161	0.4220	0.8159	0.9033
No log	53.0	159	0.7614	0.4350	0.7611	0.8724
No log	54.0	162	0.7992	0.4471	0.7989	0.8938
No log	55.0	165	0.8089	0.4656	0.8086	0.8992
No log	56.0	168	0.7920	0.4535	0.7917	0.8898
No log	57.0	171	0.7775	0.4577	0.7773	0.8816
No log	58.0	174	0.8217	0.4836	0.8215	0.9064
No log	59.0	177	0.8677	0.4572	0.8674	0.9313
No log	60.0	180	0.8936	0.4263	0.8934	0.9452
No log	61.0	183	0.7678	0.4649	0.7676	0.8761
No log	62.0	186	0.8612	0.4328	0.8611	0.9279
No log	63.0	189	0.7843	0.4530	0.7841	0.8855
No log	64.0	192	0.7769	0.4586	0.7767	0.8813
No log	65.0	195	0.8668	0.3879	0.8667	0.9310
No log	66.0	198	0.7654	0.4019	0.7653	0.8748
No log	67.0	201	0.7659	0.4089	0.7657	0.8750
No log	68.0	204	0.8756	0.4037	0.8754	0.9357
No log	69.0	207	0.7959	0.4408	0.7957	0.8920
No log	70.0	210	0.7913	0.4452	0.7911	0.8894
No log	71.0	213	0.9172	0.3944	0.9170	0.9576
No log	72.0	216	0.8429	0.4389	0.8427	0.9180
No log	73.0	219	0.8011	0.4491	0.8009	0.8949
No log	74.0	222	0.8162	0.4508	0.8160	0.9033
No log	75.0	225	0.8003	0.4412	0.8001	0.8945
No log	76.0	228	0.7787	0.4240	0.7785	0.8823
No log	77.0	231	0.8450	0.3961	0.8448	0.9191
No log	78.0	234	0.8353	0.3751	0.8351	0.9138

Framework versions

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