m-minilm-l12-h384-mal-fake-news-classification-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: 1.8794
• Accuracy: 0.4363
• F1: 0.2166

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: 256
• eval_batch_size: 256
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 9

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
1.3471	0.1639	10	1.4536	0.1553	0.0672
1.2911	0.3279	20	1.2582	0.2411	0.1537
1.2636	0.4918	30	1.3826	0.1537	0.0745
1.2154	0.6557	40	1.0301	0.6137	0.2469
1.1791	0.8197	50	1.1587	0.1574	0.0691
1.2399	0.9836	60	1.4052	0.1532	0.0720
1.1257	1.1475	70	1.1222	0.1816	0.0882
1.0962	1.3115	80	1.1921	0.3526	0.1984
1.0212	1.4754	90	1.0585	0.5495	0.2286
1.0066	1.6393	100	1.2323	0.4032	0.2104
1.0602	1.8033	110	1.1407	0.4689	0.2245
1.011	1.9672	120	1.2531	0.4574	0.2325
0.9486	2.1311	130	1.1423	0.5653	0.2472
0.9424	2.2951	140	1.3199	0.3037	0.1727
0.9735	2.4590	150	1.0358	0.6553	0.2457
0.9892	2.6230	160	1.0562	0.6295	0.2448
0.8846	2.7869	170	1.0899	0.5789	0.2376
0.9204	2.9508	180	1.3385	0.4458	0.2259
0.8902	3.1148	190	1.4512	0.4105	0.2182
0.8491	3.2787	200	1.1497	0.5447	0.2312
0.8567	3.4426	210	1.0907	0.5968	0.2363
0.8506	3.6066	220	1.3783	0.4337	0.2207
0.842	3.7705	230	1.4938	0.4095	0.2135
0.7925	3.9344	240	1.3065	0.4605	0.2154
0.7968	4.0984	250	1.2324	0.5642	0.2412
0.7923	4.2623	260	1.5467	0.3568	0.1892
0.7458	4.4262	270	1.3496	0.4816	0.2318
0.7178	4.5902	280	1.2238	0.5474	0.2347
0.7137	4.7541	290	1.2769	0.5163	0.2291
0.6994	4.9180	300	1.5223	0.4337	0.2211
0.6336	5.0820	310	1.4191	0.4732	0.2244
0.6684	5.2459	320	1.7668	0.3605	0.1954
0.6454	5.4098	330	1.4999	0.4668	0.2214
0.6323	5.5738	340	1.5826	0.4526	0.2260
0.6324	5.7377	350	1.7798	0.3837	0.2044
0.6381	5.9016	360	1.4990	0.4853	0.2250
0.5801	6.0656	370	1.7669	0.4147	0.2111
0.5732	6.2295	380	1.4530	0.5058	0.2196
0.5559	6.3934	390	1.7865	0.4063	0.2089
0.5755	6.5574	400	1.7828	0.43	0.2165
0.6038	6.7213	410	1.9098	0.4074	0.2151
0.6015	6.8852	420	1.8687	0.4063	0.2085
0.5537	7.0492	430	1.7365	0.4521	0.2224
0.5531	7.2131	440	1.6996	0.4574	0.2212
0.4868	7.3770	450	1.8164	0.4311	0.2165
0.5178	7.5410	460	1.7618	0.4474	0.2156
0.513	7.7049	470	1.7670	0.45	0.2168
0.5172	7.8689	480	1.7683	0.4589	0.2211
0.4938	8.0328	490	1.8544	0.4316	0.2116
0.4623	8.1967	500	1.7937	0.4484	0.2167
0.4679	8.3607	510	1.7579	0.4732	0.2242
0.4872	8.5246	520	1.7625	0.47	0.2233
0.4853	8.6885	530	1.8132	0.4537	0.2200
0.4772	8.8525	540	1.8794	0.4363	0.2166

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1