m-minilm-l12-h384-dra-tam-mal-meso-meme-detection

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: 0.6007
• Accuracy: 0.7117
• F1: 0.5328

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
0.6781	0.1429	2	0.6605	0.6914	0.0
0.6443	0.2857	4	0.6301	0.6914	0.0
0.6228	0.4286	6	0.6174	0.6914	0.0
0.6402	0.5714	8	0.6174	0.6914	0.0
0.6163	0.7143	10	0.6154	0.6914	0.0
0.6341	0.8571	12	0.6053	0.6914	0.0
0.5962	1.0	14	0.5972	0.6914	0.0
0.5522	1.1429	16	0.5845	0.7027	0.0704
0.577	1.2857	18	0.5854	0.7050	0.4696
0.5719	1.4286	20	0.5862	0.7117	0.4711
0.5876	1.5714	22	0.6007	0.7005	0.4981
0.5926	1.7143	24	0.5921	0.7117	0.4286
0.5405	1.8571	26	0.6141	0.6937	0.4963
0.4942	2.0	28	0.6078	0.7050	0.4609
0.5655	2.1429	30	0.6009	0.7140	0.4253
0.5051	2.2857	32	0.6054	0.7005	0.4242
0.5001	2.4286	34	0.6251	0.6892	0.4889
0.577	2.5714	36	0.6136	0.6982	0.4766
0.5034	2.7143	38	0.5884	0.7252	0.4352
0.5306	2.8571	40	0.5927	0.7252	0.3370
0.5255	3.0	42	0.5869	0.7275	0.3920
0.4786	3.1429	44	0.6012	0.7095	0.4901
0.5422	3.2857	46	0.6149	0.6914	0.4982
0.5278	3.4286	48	0.5914	0.7117	0.4797
0.4985	3.5714	50	0.5752	0.7297	0.4340
0.5262	3.7143	52	0.5707	0.7365	0.4658
0.522	3.8571	54	0.5797	0.7027	0.45
0.528	4.0	56	0.5775	0.7095	0.4603
0.4883	4.1429	58	0.5714	0.7185	0.4589
0.4504	4.2857	60	0.5754	0.7275	0.4265
0.5175	4.4286	62	0.5786	0.7252	0.4135
0.4654	4.5714	64	0.5740	0.7252	0.4299
0.5111	4.7143	66	0.5812	0.7140	0.4641
0.4264	4.8571	68	0.5962	0.7095	0.5169
0.465	5.0	70	0.6007	0.7117	0.5328
0.464	5.1429	72	0.6002	0.7117	0.4961
0.3746	5.2857	74	0.5943	0.7207	0.4918
0.4431	5.4286	76	0.5853	0.7342	0.5
0.4365	5.5714	78	0.5741	0.7365	0.5021
0.4253	5.7143	80	0.5698	0.7387	0.5167
0.4767	5.8571	82	0.5682	0.7432	0.5328
0.4225	6.0	84	0.5676	0.7387	0.5285

Framework versions

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