resnet-18-v2

This model is a fine-tuned version of microsoft/resnet-18 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.0913
Accuracy: 0.9791
Precision: 0.9943
Recall: 0.9603
F1: 0.9770
Tp: 1573
Tn: 1901
Fp: 9
Fn: 65

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: 0.003
train_batch_size: 256
eval_batch_size: 256
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: 27
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Tp	Tn	Fp	Fn
0.7107	0.0893	5	2.0020	0.4648	0.4630	0.9982	0.6326	1635	14	1896	3
0.3476	0.1786	10	1.5918	0.5417	0.5019	0.9890	0.6658	1620	302	1608	18
0.2818	0.2679	15	39.4269	0.4617	0.4617	1.0	0.6317	1638	0	1910	0
0.2810	0.3571	20	10.4487	0.4622	0.4619	1.0	0.6319	1638	2	1908	0
0.2768	0.4464	25	27.6329	0.4665	0.4639	1.0	0.6338	1638	17	1893	0
0.2856	0.5357	30	38.1928	0.4617	0.4617	1.0	0.6317	1638	0	1910	0
0.2951	0.625	35	7.1442	0.4760	0.4684	1.0	0.6380	1638	51	1859	0
0.2547	0.7143	40	0.5535	0.7359	0.6449	0.9524	0.7690	1560	1051	859	78
0.2552	0.8036	45	0.5621	0.7782	0.6980	0.9158	0.7922	1500	1261	649	138
0.2566	0.8929	50	16.2855	0.4814	0.4709	0.9982	0.6399	1635	73	1837	3
0.2710	0.9821	55	53.1862	0.5048	0.4824	0.9969	0.6502	1633	158	1752	5
0.2872	1.0714	60	2.2533	0.6040	0.5392	0.9774	0.6950	1601	542	1368	37
0.2563	1.1607	65	0.4856	0.6869	0.5999	0.9658	0.7401	1582	855	1055	56
0.2810	1.25	70	2.3695	0.7435	0.6470	0.9780	0.7788	1602	1036	874	36
0.2504	1.3393	75	6.0950	0.6387	0.5621	0.9841	0.7155	1612	654	1256	26
0.2510	1.4286	80	2.0776	0.6697	0.5902	0.9304	0.7223	1524	852	1058	114
0.2610	1.5179	85	0.2232	0.9470	0.9647	0.9188	0.9412	1505	1855	55	133
0.2148	1.6071	90	0.2034	0.9490	0.9412	0.9487	0.9450	1554	1813	97	84
0.2570	1.6964	95	0.1941	0.9479	0.9764	0.9090	0.9415	1489	1874	36	149
0.2552	1.7857	100	0.2181	0.9284	0.9402	0.9023	0.9209	1478	1816	94	160
0.2490	1.875	105	0.1679	0.9600	0.9863	0.9261	0.9553	1517	1889	21	121
0.2359	1.9643	110	0.2625	0.8923	0.9953	0.7705	0.8685	1262	1904	6	376
0.2059	2.0536	115	0.2404	0.8991	0.8851	0.8980	0.8915	1471	1719	191	167
0.2431	2.1429	120	0.2206	0.9332	0.9089	0.9505	0.9293	1557	1754	156	81
0.2064	2.2321	125	0.1875	0.9470	0.9362	0.9499	0.9430	1556	1804	106	82
0.2314	2.3214	130	0.2592	0.8853	0.8221	0.9591	0.8853	1571	1570	340	67
0.2344	2.4107	135	0.1770	0.9366	0.9144	0.9518	0.9327	1559	1764	146	79
0.2076	2.5	140	0.1330	0.9662	0.9948	0.9316	0.9622	1526	1902	8	112
0.2020	2.5893	145	0.1318	0.9603	0.9740	0.9389	0.9562	1538	1869	41	100
0.2364	2.6786	150	0.1759	0.9501	0.9980	0.8938	0.9430	1464	1907	3	174
0.2142	2.7679	155	0.2030	0.9338	0.9692	0.8846	0.9250	1449	1864	46	189
0.1821	2.8571	160	0.2203	0.9239	0.9824	0.8504	0.9116	1393	1885	25	245
0.2195	2.9464	165	0.1562	0.9493	0.9752	0.9133	0.9433	1496	1872	38	142
0.2055	3.0357	170	0.1884	0.9383	0.9201	0.9487	0.9342	1554	1775	135	84
0.2059	3.125	175	0.1479	0.9586	0.9692	0.9402	0.9544	1540	1861	49	98
0.2372	3.2143	180	0.1852	0.9357	0.9937	0.8663	0.9256	1419	1901	9	219
0.2201	3.3036	185	0.1640	0.9422	0.9350	0.9402	0.9376	1540	1803	107	98
0.1928	3.3929	190	0.1131	0.9715	0.9987	0.9396	0.9682	1539	1908	2	99
0.1767	3.4821	195	0.1428	0.9560	0.9933	0.9109	0.9503	1492	1900	10	146
0.1793	3.5714	200	0.1279	0.9648	0.9822	0.9408	0.9610	1541	1882	28	97
0.1811	3.6607	205	0.1331	0.9687	0.9805	0.9512	0.9656	1558	1879	31	80
0.2027	3.75	210	0.1223	0.9769	0.9930	0.9567	0.9745	1567	1899	11	71
0.1909	3.8393	215	0.1186	0.9777	0.9912	0.9603	0.9755	1573	1896	14	65
0.1752	3.9286	220	0.1013	0.9783	0.9943	0.9585	0.9761	1570	1901	9	68
0.2004	4.0179	225	0.0976	0.9741	0.9968	0.9469	0.9712	1551	1905	5	87
0.1840	4.1071	230	0.0937	0.9772	0.9863	0.9640	0.9750	1579	1888	22	59
0.1981	4.1964	235	0.1046	0.9735	0.9765	0.9658	0.9711	1582	1872	38	56
0.2001	4.2857	240	0.1000	0.9749	0.9814	0.9640	0.9726	1579	1880	30	59
0.1901	4.375	245	0.1042	0.9741	0.9783	0.9652	0.9717	1581	1875	35	57
0.1852	4.4643	250	0.1168	0.9690	0.9653	0.9676	0.9665	1585	1853	57	53
0.1610	4.5536	255	0.1009	0.9772	0.9869	0.9634	0.9750	1578	1889	21	60
0.1949	4.6429	260	0.1034	0.9780	0.9899	0.9621	0.9759	1576	1894	16	62
0.1570	4.7321	265	0.1029	0.9794	0.9912	0.9640	0.9774	1579	1896	14	59
0.1727	4.8214	270	0.0970	0.9797	0.9943	0.9615	0.9777	1575	1901	9	63
0.1484	4.9107	275	0.0943	0.9797	0.9956	0.9603	0.9776	1573	1903	7	65
0.1714	5.0	280	0.0913	0.9791	0.9943	0.9603	0.9770	1573	1901	9	65

Framework versions

Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 4.0.0
Tokenizers 0.22.2

waelhasan/resnet-18-v2

resnet-18-v2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions