Modelpublic

genki10/ASAP_FineTuningBERT_UnAugV7_k1_task1_organization_k1_k1_fold4

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes1downloads

Model Card

ASAPFineTuningBERTUnAugV7k1task1organizationk1k1fold4

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: 0.8479
—Qwk: 0.6040
—Mse: 0.8479
—Rmse: 0.9208

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	12.4120	0.0027	12.4120	3.5231
No log	2.0	2	10.7625	0.0134	10.7625	3.2806
No log	3.0	3	10.1904	0.0018	10.1904	3.1922
No log	4.0	4	9.7189	0.0018	9.7189	3.1175
No log	5.0	5	9.2588	0.0018	9.2588	3.0428
No log	6.0	6	8.8692	0.0018	8.8692	2.9781
No log	7.0	7	8.5366	0.0	8.5366	2.9217
No log	8.0	8	8.1969	0.0	8.1969	2.8630
No log	9.0	9	7.7776	0.0	7.7776	2.7888
No log	10.0	10	7.2121	0.0	7.2121	2.6855
No log	11.0	11	6.5103	0.0	6.5103	2.5515
No log	12.0	12	5.7691	0.0190	5.7691	2.4019
No log	13.0	13	5.2658	0.0176	5.2658	2.2947
No log	14.0	14	4.7218	0.0040	4.7218	2.1730
No log	15.0	15	4.1371	0.0040	4.1371	2.0340
No log	16.0	16	3.7517	0.0040	3.7517	1.9369
No log	17.0	17	3.4886	0.0040	3.4886	1.8678
No log	18.0	18	3.0870	0.0040	3.0870	1.7570
No log	19.0	19	2.5764	-0.0006	2.5764	1.6051
No log	20.0	20	2.1386	0.1111	2.1386	1.4624
No log	21.0	21	1.8224	0.0445	1.8224	1.3500
No log	22.0	22	1.6061	0.0420	1.6061	1.2673
No log	23.0	23	1.4381	0.0420	1.4381	1.1992
No log	24.0	24	1.3381	0.0420	1.3381	1.1568
No log	25.0	25	1.2815	0.0420	1.2815	1.1320
No log	26.0	26	1.1965	0.0420	1.1965	1.0938
No log	27.0	27	1.0834	0.0420	1.0834	1.0409
No log	28.0	28	0.9810	0.0420	0.9810	0.9905
No log	29.0	29	0.8904	0.1102	0.8904	0.9436
No log	30.0	30	0.8524	0.2177	0.8524	0.9232
No log	31.0	31	0.8164	0.3244	0.8164	0.9035
No log	32.0	32	0.8378	0.3548	0.8378	0.9153
No log	33.0	33	0.8740	0.3443	0.8740	0.9349
No log	34.0	34	0.8337	0.3876	0.8337	0.9131
No log	35.0	35	0.7724	0.4446	0.7724	0.8789
No log	36.0	36	0.7007	0.4774	0.7007	0.8371
No log	37.0	37	0.7635	0.4766	0.7635	0.8738
No log	38.0	38	0.9426	0.4335	0.9426	0.9709
No log	39.0	39	1.0714	0.3958	1.0714	1.0351
No log	40.0	40	0.9254	0.4650	0.9254	0.9620
No log	41.0	41	0.9768	0.4572	0.9768	0.9883
No log	42.0	42	1.2373	0.3794	1.2373	1.1123
No log	43.0	43	1.2191	0.3847	1.2191	1.1041
No log	44.0	44	1.0573	0.4700	1.0573	1.0282
No log	45.0	45	0.9150	0.5036	0.9150	0.9566
No log	46.0	46	0.8637	0.5195	0.8637	0.9293
No log	47.0	47	0.9751	0.5041	0.9751	0.9875
No log	48.0	48	1.3711	0.4152	1.3711	1.1709
No log	49.0	49	1.5056	0.3625	1.5056	1.2270
No log	50.0	50	1.2979	0.4548	1.2979	1.1393
No log	51.0	51	0.9740	0.5159	0.9740	0.9869
No log	52.0	52	0.8307	0.5977	0.8307	0.9114
No log	53.0	53	0.8565	0.5835	0.8565	0.9255
No log	54.0	54	0.8510	0.5890	0.8510	0.9225
No log	55.0	55	0.8217	0.5837	0.8217	0.9065
No log	56.0	56	0.9129	0.5383	0.9129	0.9555
No log	57.0	57	1.0323	0.5264	1.0323	1.0160
No log	58.0	58	1.0543	0.5213	1.0543	1.0268
No log	59.0	59	0.9750	0.5249	0.9750	0.9874
No log	60.0	60	0.8822	0.5611	0.8822	0.9393
No log	61.0	61	0.8536	0.5926	0.8536	0.9239
No log	62.0	62	0.8537	0.6095	0.8537	0.9240
No log	63.0	63	0.8274	0.6130	0.8274	0.9096
No log	64.0	64	0.7758	0.5956	0.7758	0.8808
No log	65.0	65	0.7778	0.5661	0.7778	0.8819
No log	66.0	66	0.8111	0.5617	0.8111	0.9006
No log	67.0	67	0.8172	0.5519	0.8172	0.9040
No log	68.0	68	0.7924	0.5660	0.7924	0.8902
No log	69.0	69	0.7794	0.5928	0.7794	0.8828
No log	70.0	70	0.8230	0.5993	0.8230	0.9072
No log	71.0	71	0.8688	0.6112	0.8688	0.9321
No log	72.0	72	0.8738	0.6024	0.8738	0.9348
No log	73.0	73	0.8479	0.6040	0.8479	0.9208

Framework versions

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