Modelpublic

tringuyen-uit/VP_ViSoBERT_syl_ViWikiFC

sourceHugging Faceupdated 2y agoView on Hugging Face

0likes8downloads

Model Card

VPViSoBERTsyl_ViWikiFC

This model is a fine-tuned version of tringuyen-uit/VP_ViSoBERT_syl_ViWikiFC on the None dataset. It achieves the following results on the evaluation set:

—Loss: 1.1555
—Accuracy: 0.6445

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: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6994	0.05	100	0.9688	0.6158
0.6904	0.1	200	0.9753	0.6014
0.7969	0.14	300	0.9446	0.5871
0.6801	0.19	400	0.9912	0.6057
0.7089	0.24	500	0.9617	0.5861
0.6627	0.29	600	1.0585	0.5689
0.6792	0.33	700	1.0064	0.6230
0.6702	0.38	800	1.0593	0.5818
0.6252	0.43	900	0.9621	0.5967
0.6262	0.48	1000	1.0152	0.5957
0.6515	0.53	1100	0.9539	0.6225
0.6596	0.57	1200	0.9188	0.6067
0.6458	0.62	1300	0.9318	0.6201
0.6087	0.67	1400	0.9532	0.6172
0.6282	0.72	1500	1.0107	0.6244
0.6266	0.76	1600	1.0199	0.6096
0.6165	0.81	1700	1.0973	0.6096
0.5869	0.86	1800	0.9177	0.6325
0.596	0.91	1900	0.8821	0.6364
0.6073	0.96	2000	0.9350	0.6306
0.5921	1.0	2100	0.9606	0.6282
0.4551	1.05	2200	1.0386	0.6373
0.3922	1.1	2300	1.1936	0.6368
0.39	1.15	2400	1.1922	0.6316
0.442	1.19	2500	1.1599	0.6220
0.4092	1.24	2600	1.3106	0.6196
0.4582	1.29	2700	1.1817	0.6316
0.4356	1.34	2800	1.1257	0.6316
0.4145	1.39	2900	1.1899	0.6354
0.4379	1.43	3000	1.1385	0.6388
0.4222	1.48	3100	1.1844	0.6249
0.3758	1.53	3200	1.2444	0.6311
0.4114	1.58	3300	1.1908	0.6349
0.4449	1.62	3400	1.1483	0.6273
0.4046	1.67	3500	1.1977	0.6306
0.4274	1.72	3600	1.1520	0.6450
0.3785	1.77	3700	1.1665	0.6330
0.3854	1.82	3800	1.1680	0.6474
0.3562	1.86	3900	1.1616	0.6459
0.3938	1.91	4000	1.1823	0.6397
0.5083	1.96	4100	1.1555	0.6445

Framework versions

—Transformers 4.39.3
—Pytorch 2.1.2
—Datasets 2.18.0
—Tokenizers 0.15.2