livinNector/m-minilm-l12-h384-dra-tam-mal-ai-gen-review-classification-finetune-2

m-minilm-l12-h384-dra-tam-mal-ai-gen-review-classification-head-finetune

This model is a fine-tuned version of microsoft/Multilingual-MiniLM-L12-H384 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1110
• Accuracy: 0.9814
• F1: 0.9815

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: 0.0001
• train_batch_size: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.6933	0.1818	2	0.6901	0.5	0.6667
0.6887	0.3636	4	0.6762	0.5	0.6667
0.6755	0.5455	6	0.6341	0.9317	0.9345
0.6281	0.7273	8	0.5641	0.9286	0.9329
0.5497	0.9091	10	0.4588	0.9472	0.9477
0.4924	1.0909	12	0.3883	0.9534	0.9541
0.368	1.2727	14	0.3217	0.9627	0.9632
0.3358	1.4545	16	0.2773	0.9596	0.9595
0.2909	1.6364	18	0.2571	0.9503	0.9524
0.2534	1.8182	20	0.2095	0.9752	0.9756
0.2041	2.0	22	0.1872	0.9783	0.9786
0.202	2.1818	24	0.1726	0.9783	0.9785
0.2001	2.3636	26	0.1768	0.9689	0.9697
0.1821	2.5455	28	0.1644	0.9720	0.9728
0.1888	2.7273	30	0.1535	0.9752	0.9752
0.1852	2.9091	32	0.1519	0.9720	0.9716
0.1631	3.0909	34	0.1502	0.9720	0.9716
0.1592	3.2727	36	0.1332	0.9783	0.9785
0.1199	3.4545	38	0.1367	0.9720	0.9726
0.1984	3.6364	40	0.2671	0.9193	0.9253
0.2559	3.8182	42	0.2112	0.9379	0.9415
0.1334	4.0	44	0.1488	0.9658	0.9670
0.1933	4.1818	46	0.1330	0.9752	0.9756
0.1637	4.3636	48	0.1213	0.9783	0.9786
0.1183	4.5455	50	0.1203	0.9783	0.9785
0.1272	4.7273	52	0.1155	0.9783	0.9785
0.1024	4.9091	54	0.1110	0.9814	0.9815
0.0713	5.0909	56	0.1101	0.9814	0.9815
0.1423	5.2727	58	0.1102	0.9814	0.9815
0.1117	5.4545	60	0.1090	0.9814	0.9815
0.1044	5.6364	62	0.1086	0.9814	0.9815
0.1205	5.8182	64	0.1083	0.9814	0.9815
0.0593	6.0	66	0.1080	0.9814	0.9815

Framework versions

• Transformers 4.45.2
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.20.3