Modelpublic

genki10/BERT_V8_sp10_lw20_ex100_lo100_k5_k5_fold1

sourceHugging Faceapache-2.0updated 1y agoView on Hugging Face

0likes6downloads

Model Card

BERTV8sp10lw20ex100lo100k5k5fold1

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: 0.5772
—Qwk: 0.5842
—Mse: 0.5760
—Rmse: 0.7590

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
—trainbatchsize: 64
—evalbatchsize: 64
—seed: 42
—optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	1.0	3	8.4645	0.0	8.4621	2.9090
No log	2.0	6	6.1646	0.0	6.1624	2.4824
No log	3.0	9	4.5537	0.0	4.5515	2.1334
No log	4.0	12	3.2625	0.0	3.2605	1.8057
No log	5.0	15	2.1782	0.1667	2.1764	1.4753
No log	6.0	18	1.4487	0.0	1.4472	1.2030
No log	7.0	21	1.0478	0.0	1.0464	1.0230
No log	8.0	24	0.8526	0.3845	0.8513	0.9227
No log	9.0	27	0.7383	0.3239	0.7372	0.8586
No log	10.0	30	0.6611	0.3493	0.6601	0.8125
No log	11.0	33	0.8173	0.3692	0.8160	0.9033
No log	12.0	36	0.5889	0.4040	0.5881	0.7669
No log	13.0	39	1.0005	0.3159	0.9991	0.9996
No log	14.0	42	0.5512	0.5281	0.5502	0.7418
No log	15.0	45	0.7687	0.4478	0.7675	0.8761
No log	16.0	48	0.5171	0.6125	0.5163	0.7185
No log	17.0	51	0.6203	0.5621	0.6192	0.7869
No log	18.0	54	0.5285	0.6084	0.5275	0.7263
No log	19.0	57	0.5618	0.6054	0.5608	0.7489
No log	20.0	60	0.5309	0.6454	0.5300	0.7280
No log	21.0	63	0.6605	0.5515	0.6594	0.8121
No log	22.0	66	0.5928	0.6214	0.5918	0.7693
No log	23.0	69	0.6201	0.6089	0.6189	0.7867
No log	24.0	72	0.6099	0.6085	0.6088	0.7802
No log	25.0	75	0.6878	0.5107	0.6865	0.8285
No log	26.0	78	0.5950	0.5661	0.5940	0.7707
No log	27.0	81	0.6088	0.5531	0.6075	0.7794
No log	28.0	84	0.5762	0.6140	0.5750	0.7583
No log	29.0	87	0.6238	0.6049	0.6226	0.7891
No log	30.0	90	0.6827	0.5464	0.6813	0.8254
No log	31.0	93	0.6405	0.5531	0.6392	0.7995
No log	32.0	96	0.5923	0.5736	0.5911	0.7688
No log	33.0	99	0.7023	0.5418	0.7009	0.8372
No log	34.0	102	0.5642	0.6098	0.5630	0.7504
No log	35.0	105	0.6244	0.5730	0.6231	0.7894
No log	36.0	108	0.5943	0.6043	0.5932	0.7702
No log	37.0	111	0.6368	0.5394	0.6355	0.7972
No log	38.0	114	0.5724	0.5908	0.5712	0.7558
No log	39.0	117	0.5822	0.5789	0.5810	0.7622
No log	40.0	120	0.5721	0.5924	0.5709	0.7556
No log	41.0	123	0.6190	0.5753	0.6178	0.7860
No log	42.0	126	0.6006	0.5739	0.5995	0.7743
No log	43.0	129	0.6224	0.5465	0.6212	0.7882
No log	44.0	132	0.5914	0.5672	0.5903	0.7683
No log	45.0	135	0.6374	0.5685	0.6360	0.7975
No log	46.0	138	0.6095	0.5783	0.6083	0.7799
No log	47.0	141	0.6263	0.5791	0.6249	0.7905
No log	48.0	144	0.5868	0.5926	0.5855	0.7652
No log	49.0	147	0.6410	0.5380	0.6397	0.7998
No log	50.0	150	0.5772	0.5842	0.5760	0.7590

Framework versions

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