mobilenet_v4_small

This model is a fine-tuned version of timm/mobilenetv4_conv_small.e2400_r224_in1k on an unknown dataset. It achieves the following results on the evaluation set:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 0.0001
train_batch_size: 128
eval_batch_size: 128
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 220
num_epochs: 20
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Tp	Tn	Fp	Fn
0.2458	1.0	111	0.3615	0.8393	0.7651	0.9408	0.8439	1541	1437	473	97
0.2261	2.0	222	0.2836	0.8715	0.8050	0.9524	0.8725	1560	1532	378	78
0.2261	3.0	333	0.3670	0.8089	0.7164	0.9701	0.8242	1589	1281	629	49
0.2150	4.0	444	0.1581	0.9532	0.9450	0.9542	0.9496	1563	1819	91	75
0.1853	5.0	555	0.1492	0.9555	0.9447	0.9597	0.9522	1572	1818	92	66
0.1509	6.0	666	0.1260	0.9645	0.9621	0.9609	0.9615	1574	1848	62	64
0.1568	7.0	777	0.0952	0.9777	0.9821	0.9695	0.9757	1588	1881	29	50
0.1286	8.0	888	0.0878	0.9755	0.9844	0.9621	0.9731	1576	1885	25	62
0.1742	9.0	999	0.0950	0.9744	0.9737	0.9707	0.9722	1590	1867	43	48
0.1623	10.0	1110	0.0944	0.9721	0.9718	0.9676	0.9697	1585	1864	46	53
0.1550	11.0	1221	0.0835	0.9808	0.9931	0.9652	0.9789	1581	1899	11	57
0.1526	12.0	1332	0.1751	0.9459	0.9141	0.9744	0.9433	1596	1760	150	42
0.1175	13.0	1443	0.0638	0.9834	0.9925	0.9713	0.9818	1591	1898	12	47
0.1349	14.0	1554	0.0754	0.9794	0.9839	0.9713	0.9776	1591	1884	26	47
0.1121	15.0	1665	0.0822	0.9794	0.9851	0.9701	0.9775	1589	1886	24	49
0.1319	16.0	1776	0.0747	0.9808	0.9834	0.9750	0.9792	1597	1883	27	41
0.1367	17.0	1887	0.0646	0.9828	0.9907	0.9719	0.9812	1592	1895	15	46
0.1297	18.0	1998	0.0690	0.9820	0.9858	0.9750	0.9804	1597	1887	23	41
0.1218	19.0	2109	0.0692	0.9811	0.9864	0.9725	0.9794	1593	1888	22	45
0.1380	20.0	2220	0.0668	0.9817	0.9870	0.9731	0.9800	1594	1889	21	44