Modelpublic

genki10/ASAP_FineTuningBERT_UnAugV6_k1_task1_organization_k1_k1_fold4

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes1downloads

Model Card

ASAPFineTuningBERTUnAugV6k1task1organizationk1k1fold4

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: 0.8487
—Qwk: 0.5980
—Mse: 0.8487
—Rmse: 0.9212

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	12.3540	-0.0037	12.3540	3.5148
No log	2.0	2	11.1609	-0.0241	11.1609	3.3408
No log	3.0	3	9.7667	0.0018	9.7667	3.1252
No log	4.0	4	8.6627	0.0018	8.6627	2.9433
No log	5.0	5	7.7487	0.0018	7.7487	2.7837
No log	6.0	6	6.9227	0.0018	6.9227	2.6311
No log	7.0	7	6.2576	0.0018	6.2576	2.5015
No log	8.0	8	5.5544	0.0162	5.5544	2.3568
No log	9.0	9	4.8655	0.0138	4.8655	2.2058
No log	10.0	10	4.3021	0.0079	4.3021	2.0742
No log	11.0	11	3.7928	0.0040	3.7928	1.9475
No log	12.0	12	3.3353	0.0040	3.3353	1.8263
No log	13.0	13	2.9305	0.0040	2.9305	1.7119
No log	14.0	14	2.6226	0.0040	2.6226	1.6195
No log	15.0	15	2.3390	0.1062	2.3390	1.5294
No log	16.0	16	2.1040	0.1703	2.1040	1.4505
No log	17.0	17	1.9080	0.0923	1.9080	1.3813
No log	18.0	18	1.6920	0.0445	1.6920	1.3008
No log	19.0	19	1.5269	0.0445	1.5269	1.2357
No log	20.0	20	1.3951	0.0445	1.3951	1.1811
No log	21.0	21	1.2675	0.0445	1.2675	1.1258
No log	22.0	22	1.1831	0.0547	1.1831	1.0877
No log	23.0	23	1.0983	0.0710	1.0983	1.0480
No log	24.0	24	0.9900	0.0547	0.9900	0.9950
No log	25.0	25	0.9162	0.0547	0.9162	0.9572
No log	26.0	26	0.8552	0.0919	0.8552	0.9248
No log	27.0	27	0.8040	0.2784	0.8040	0.8967
No log	28.0	28	0.7601	0.4600	0.7601	0.8718
No log	29.0	29	0.6933	0.5524	0.6933	0.8327
No log	30.0	30	0.6447	0.5338	0.6447	0.8029
No log	31.0	31	0.6113	0.5624	0.6113	0.7819
No log	32.0	32	0.5846	0.5735	0.5846	0.7646
No log	33.0	33	0.5664	0.5348	0.5664	0.7526
No log	34.0	34	0.5416	0.5441	0.5416	0.7359
No log	35.0	35	0.5407	0.5940	0.5407	0.7353
No log	36.0	36	0.5160	0.5723	0.5160	0.7183
No log	37.0	37	0.5464	0.5048	0.5464	0.7392
No log	38.0	38	0.5635	0.4993	0.5635	0.7507
No log	39.0	39	0.5088	0.5516	0.5088	0.7133
No log	40.0	40	0.4949	0.5614	0.4949	0.7035
No log	41.0	41	0.5201	0.5486	0.5201	0.7211
No log	42.0	42	0.5689	0.5264	0.5689	0.7542
No log	43.0	43	0.5564	0.5389	0.5564	0.7460
No log	44.0	44	0.5533	0.5897	0.5533	0.7438
No log	45.0	45	0.5704	0.6046	0.5704	0.7553
No log	46.0	46	0.6580	0.5561	0.6580	0.8112
No log	47.0	47	0.6458	0.5798	0.6458	0.8036
No log	48.0	48	0.5852	0.6218	0.5852	0.7650
No log	49.0	49	0.5765	0.6264	0.5765	0.7592
No log	50.0	50	0.6424	0.6067	0.6424	0.8015
No log	51.0	51	0.7589	0.5623	0.7589	0.8712
No log	52.0	52	0.7443	0.5725	0.7443	0.8627
No log	53.0	53	0.6016	0.6359	0.6016	0.7756
No log	54.0	54	0.5784	0.6353	0.5784	0.7605
No log	55.0	55	0.6759	0.6053	0.6759	0.8221
No log	56.0	56	0.7824	0.5802	0.7824	0.8845
No log	57.0	57	0.8113	0.5679	0.8113	0.9007
No log	58.0	58	0.6990	0.6087	0.6990	0.8361
No log	59.0	59	0.6992	0.6079	0.6992	0.8362
No log	60.0	60	0.8220	0.5655	0.8220	0.9067
No log	61.0	61	0.9930	0.5108	0.9930	0.9965
No log	62.0	62	0.9825	0.5299	0.9825	0.9912
No log	63.0	63	0.8659	0.5631	0.8659	0.9305
No log	64.0	64	0.7615	0.6099	0.7615	0.8726
No log	65.0	65	0.7747	0.6050	0.7747	0.8802
No log	66.0	66	0.9046	0.5770	0.9046	0.9511
No log	67.0	67	1.1602	0.4883	1.1602	1.0771
No log	68.0	68	1.2434	0.4737	1.2434	1.1151
No log	69.0	69	1.1567	0.5080	1.1567	1.0755
No log	70.0	70	0.9457	0.5777	0.9457	0.9725
No log	71.0	71	0.8041	0.6117	0.8041	0.8967
No log	72.0	72	0.7807	0.6114	0.7807	0.8836
No log	73.0	73	0.8487	0.5980	0.8487	0.9212

Framework versions

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