Version22NewTestASAP_FineTuningBERT_AugV22_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: 0.5424
• Qwk: 0.6166
• Mse: 0.5416
• Rmse: 0.7360

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	9.1974	0.0	9.1948	3.0323
No log	2.0	6	6.4925	0.0	6.4899	2.5475
No log	3.0	9	4.4360	0.0079	4.4339	2.1057
No log	4.0	12	3.0837	0.0	3.0818	1.7555
No log	5.0	15	2.0512	0.0759	2.0495	1.4316
No log	6.0	18	1.4735	0.0211	1.4720	1.2133
No log	7.0	21	1.1493	0.0106	1.1479	1.0714
No log	8.0	24	1.2815	0.0521	1.2799	1.1313
No log	9.0	27	1.3040	0.0894	1.3024	1.1412
No log	10.0	30	0.9186	0.0867	0.9172	0.9577
No log	11.0	33	1.1277	0.1368	1.1262	1.0612
No log	12.0	36	0.8389	0.2688	0.8374	0.9151
No log	13.0	39	0.9294	0.2271	0.9280	0.9633
No log	14.0	42	0.7591	0.3566	0.7578	0.8705
No log	15.0	45	0.8329	0.3268	0.8317	0.9120
No log	16.0	48	0.5999	0.4745	0.5989	0.7739
No log	17.0	51	0.6572	0.4295	0.6561	0.8100
No log	18.0	54	0.6004	0.4875	0.5994	0.7742
No log	19.0	57	0.7535	0.4233	0.7525	0.8675
No log	20.0	60	0.5669	0.5158	0.5659	0.7523
No log	21.0	63	0.9299	0.3519	0.9289	0.9638
No log	22.0	66	0.5987	0.4830	0.5977	0.7731
No log	23.0	69	0.6743	0.4220	0.6733	0.8206
No log	24.0	72	0.6499	0.4569	0.6490	0.8056
No log	25.0	75	0.6877	0.4213	0.6867	0.8287
No log	26.0	78	0.6006	0.5459	0.5997	0.7744
No log	27.0	81	0.8220	0.4334	0.8210	0.9061
No log	28.0	84	0.6872	0.4643	0.6863	0.8284
No log	29.0	87	0.5866	0.5513	0.5858	0.7654
No log	30.0	90	0.5994	0.5162	0.5985	0.7736
No log	31.0	93	0.5942	0.5570	0.5933	0.7703
No log	32.0	96	0.5289	0.5784	0.5281	0.7267
No log	33.0	99	0.5470	0.5750	0.5462	0.7390
No log	34.0	102	0.5408	0.5700	0.5399	0.7348
No log	35.0	105	0.6513	0.4977	0.6505	0.8065
No log	36.0	108	0.6206	0.5411	0.6196	0.7872
No log	37.0	111	0.7203	0.4858	0.7194	0.8482
No log	38.0	114	0.6197	0.5281	0.6189	0.7867
No log	39.0	117	0.6449	0.5196	0.6441	0.8026
No log	40.0	120	0.6942	0.5263	0.6934	0.8327
No log	41.0	123	0.5474	0.575	0.5466	0.7393
No log	42.0	126	0.5713	0.5773	0.5706	0.7554
No log	43.0	129	0.8873	0.4791	0.8865	0.9415
No log	44.0	132	0.5472	0.6029	0.5464	0.7392
No log	45.0	135	0.4902	0.6300	0.4894	0.6996
No log	46.0	138	0.6420	0.5608	0.6413	0.8008
No log	47.0	141	0.5376	0.6255	0.5369	0.7327
No log	48.0	144	0.6317	0.5563	0.6310	0.7944
No log	49.0	147	0.6274	0.5548	0.6268	0.7917
No log	50.0	150	0.6554	0.5564	0.6548	0.8092
No log	51.0	153	0.5956	0.5827	0.5950	0.7713
No log	52.0	156	0.7222	0.5449	0.7216	0.8495
No log	53.0	159	0.5972	0.5820	0.5965	0.7723
No log	54.0	162	0.7195	0.5330	0.7188	0.8478
No log	55.0	165	0.7421	0.5352	0.7416	0.8612
No log	56.0	168	0.6211	0.5997	0.6205	0.7877
No log	57.0	171	0.7630	0.5460	0.7625	0.8732
No log	58.0	174	0.5582	0.6185	0.5575	0.7467
No log	59.0	177	0.6811	0.5552	0.6804	0.8249
No log	60.0	180	0.7256	0.5606	0.7249	0.8514
No log	61.0	183	0.6533	0.5606	0.6526	0.8079
No log	62.0	186	0.7551	0.5560	0.7545	0.8686
No log	63.0	189	0.6832	0.5511	0.6825	0.8261
No log	64.0	192	0.7105	0.5454	0.7098	0.8425
No log	65.0	195	0.7241	0.5371	0.7234	0.8505
No log	66.0	198	0.9346	0.4899	0.9340	0.9664
No log	67.0	201	0.8120	0.5306	0.8114	0.9008
No log	68.0	204	0.6664	0.5622	0.6657	0.8159
No log	69.0	207	0.6450	0.5724	0.6443	0.8027
No log	70.0	210	0.6091	0.5957	0.6084	0.7800
No log	71.0	213	0.7784	0.5318	0.7779	0.8820
No log	72.0	216	0.7841	0.5251	0.7835	0.8852
No log	73.0	219	0.8599	0.5018	0.8593	0.9270
No log	74.0	222	0.7060	0.5399	0.7053	0.8398
No log	75.0	225	0.5424	0.6166	0.5416	0.7360

Framework versions

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