Modelpublic

genki10/ASAP_FineTuningBERT_UnAugV6_k1_task1_organization_k1_k1_fold0

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes3downloads

Model Card

ASAPFineTuningBERTUnAugV6k1task1organizationk1k1fold0

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: 0.6735
—Qwk: 0.6057
—Mse: 0.6735
—Rmse: 0.8207

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	1	13.0791	0.0	13.0791	3.6165
No log	2.0	2	12.5724	-0.0007	12.5724	3.5458
No log	3.0	3	11.2937	0.0037	11.2937	3.3606
No log	4.0	4	10.1043	0.0050	10.1043	3.1787
No log	5.0	5	9.0024	0.0054	9.0024	3.0004
No log	6.0	6	7.6978	0.0036	7.6978	2.7745
No log	7.0	7	6.4265	0.0062	6.4265	2.5350
No log	8.0	8	5.8906	0.0032	5.8906	2.4271
No log	9.0	9	5.1660	0.0153	5.1660	2.2729
No log	10.0	10	4.0896	0.0115	4.0896	2.0223
No log	11.0	11	3.7970	0.0115	3.7970	1.9486
No log	12.0	12	3.4356	0.0115	3.4356	1.8535
No log	13.0	13	2.9491	0.0115	2.9491	1.7173
No log	14.0	14	2.4865	0.1045	2.4865	1.5769
No log	15.0	15	2.1332	0.1152	2.1332	1.4606
No log	16.0	16	2.0460	0.0848	2.0460	1.4304
No log	17.0	17	1.7066	0.0672	1.7066	1.3064
No log	18.0	18	1.5179	0.0520	1.5179	1.2320
No log	19.0	19	1.4076	0.0520	1.4076	1.1864
No log	20.0	20	1.2757	0.0419	1.2757	1.1295
No log	21.0	21	1.1168	0.0419	1.1168	1.0568
No log	22.0	22	0.9794	0.0316	0.9794	0.9896
No log	23.0	23	0.9103	0.0419	0.9103	0.9541
No log	24.0	24	0.8421	0.1540	0.8421	0.9177
No log	25.0	25	0.7712	0.4410	0.7712	0.8782
No log	26.0	26	0.7402	0.4548	0.7402	0.8604
No log	27.0	27	0.7068	0.4674	0.7068	0.8407
No log	28.0	28	0.6391	0.4659	0.6391	0.7994
No log	29.0	29	0.5827	0.4872	0.5827	0.7633
No log	30.0	30	0.5561	0.5021	0.5561	0.7457
No log	31.0	31	0.5347	0.4599	0.5347	0.7312
No log	32.0	32	0.5258	0.4655	0.5258	0.7251
No log	33.0	33	0.4946	0.5041	0.4946	0.7033
No log	34.0	34	0.5455	0.5545	0.5455	0.7386
No log	35.0	35	0.5296	0.5550	0.5296	0.7278
No log	36.0	36	0.4865	0.4990	0.4865	0.6975
No log	37.0	37	0.5350	0.4733	0.5350	0.7314
No log	38.0	38	0.5540	0.5036	0.5540	0.7443
No log	39.0	39	0.5060	0.5753	0.5060	0.7113
No log	40.0	40	0.5066	0.5881	0.5066	0.7118
No log	41.0	41	0.5172	0.6062	0.5172	0.7192
No log	42.0	42	0.6032	0.5852	0.6032	0.7766
No log	43.0	43	0.6149	0.5876	0.6149	0.7842
No log	44.0	44	0.5432	0.5982	0.5432	0.7370
No log	45.0	45	0.5558	0.5939	0.5558	0.7455
No log	46.0	46	0.6419	0.5679	0.6419	0.8012
No log	47.0	47	0.6658	0.5593	0.6658	0.8159
No log	48.0	48	0.6155	0.5812	0.6155	0.7845
No log	49.0	49	0.5522	0.6224	0.5522	0.7431
No log	50.0	50	0.5584	0.6163	0.5584	0.7473
No log	51.0	51	0.6355	0.5884	0.6355	0.7972
No log	52.0	52	0.7108	0.5590	0.7108	0.8431
No log	53.0	53	0.6738	0.5744	0.6738	0.8208
No log	54.0	54	0.6327	0.6069	0.6327	0.7954
No log	55.0	55	0.6491	0.5949	0.6491	0.8057
No log	56.0	56	0.6272	0.6086	0.6272	0.7919
No log	57.0	57	0.5987	0.6125	0.5987	0.7738
No log	58.0	58	0.6020	0.6085	0.6020	0.7759
No log	59.0	59	0.6581	0.5878	0.6581	0.8112
No log	60.0	60	0.7387	0.5545	0.7387	0.8595
No log	61.0	61	0.8453	0.5340	0.8453	0.9194
No log	62.0	62	0.7955	0.5525	0.7955	0.8919
No log	63.0	63	0.6592	0.6163	0.6592	0.8119
No log	64.0	64	0.6338	0.6092	0.6338	0.7961
No log	65.0	65	0.6669	0.6164	0.6669	0.8166
No log	66.0	66	0.7722	0.5697	0.7722	0.8788
No log	67.0	67	0.8479	0.5351	0.8479	0.9208
No log	68.0	68	0.7925	0.5451	0.7925	0.8902
No log	69.0	69	0.6735	0.6057	0.6735	0.8207

Framework versions

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