Modelpublic

slnkvdns/finetune-instance-segmentation-alpha-dent-mask2former

sourceHugging Faceupdated 4mo agoView on Hugging Face

0likes18downloads

Model Card

finetune-instance-segmentation-alpha-dent-mask2former

This model is a fine-tuned version of slnkvdns/finetune-instance-segmentation-alpha-dent-mask2former on the slnkvdns/AlphaDent dataset. It achieves the following results on the evaluation set:

—Loss: 26.6500
—Map: 0.2861
—Map 50: 0.425
—Map 75: 0.2873
—Map Small: 0.1204
—Map Medium: 0.3288
—Map Large: 0.9391
—Mar 1: 0.1955
—Mar 10: 0.3752
—Mar 100: 0.3935
—Mar Small: 0.2283
—Mar Medium: 0.4298
—Mar Large: 0.9416
—Map Background: 0.9668
—Mar 100 Background: 0.9747
—Map Abrasion: 0.659
—Mar 100 Abrasion: 0.8286
—Map Filling: 0.2331
—Mar 100 Filling: 0.3693
—Map Crown: 0.6927
—Mar 100 Crown: 0.7263
—Map Caries class 1: 0.1201
—Mar 100 Caries class 1: 0.2672
—Map Caries class 2: 0.0486
—Mar 100 Caries class 2: 0.2292
—Map Caries class 3: 0.005
—Mar 100 Caries class 3: 0.0758
—Map Caries class 4: 0.0028
—Mar 100 Caries class 4: 0.1
—Map Caries class 5: 0.1192
—Mar 100 Caries class 5: 0.2436
—Map Caries class 6: 0.0135
—Mar 100 Caries class 6: 0.12

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: 1e-05
—trainbatchsize: 4
—evalbatchsize: 8
—seed: 42
—gradientaccumulationsteps: 2
—totaltrainbatch_size: 8
—optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
—lrschedulertype: constant
—num_epochs: 10.0
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Map 50	Map 75	Map Small	Map Medium	Map Large	Mar 1	Mar 10	Mar 100	Mar Small	Mar Medium	Mar Large	Map Background	Mar 100 Background	Map Abrasion	Mar 100 Abrasion	Map Filling	Mar 100 Filling	Map Crown	Mar 100 Crown	Map Caries class 1	Mar 100 Caries class 1	Map Caries class 2	Mar 100 Caries class 2	Map Caries class 3	Mar 100 Caries class 3	Map Caries class 4	Mar 100 Caries class 4	Map Caries class 5	Mar 100 Caries class 5	Map Caries class 6	Mar 100 Caries class 6
17.9799	1.0	155	24.1491	0.2825	0.4176	0.2842	0.1173	0.3169	0.8899	0.1842	0.3561	0.3783	0.2077	0.4217	0.9416	0.9678	0.9747	0.6737	0.8454	0.2432	0.395	0.6666	0.7263	0.1127	0.2621	0.0312	0.1764	0.0071	0.1	0.0011	0.025	0.1187	0.2577	0.0035	0.02
17.2483	2.0	310	24.3290	0.2887	0.429	0.2872	0.1255	0.3259	0.9385	0.2001	0.3738	0.39	0.2269	0.4209	0.9408	0.965	0.9723	0.67	0.8471	0.2452	0.3765	0.6667	0.7211	0.1229	0.231	0.0316	0.1875	0.0055	0.0939	0.0115	0.125	0.1266	0.2654	0.0416	0.08
16.9506	3.0	465	24.6849	0.2829	0.4108	0.2832	0.1231	0.3163	0.9391	0.1777	0.3604	0.3784	0.2088	0.4175	0.9416	0.9669	0.9747	0.6682	0.8426	0.2406	0.381	0.6506	0.7263	0.1319	0.2483	0.0324	0.1972	0.0085	0.0879	0.0	0.0	0.1149	0.2462	0.0156	0.08
16.7543	4.0	620	25.2413	0.2871	0.4272	0.2895	0.1227	0.3107	0.8882	0.183	0.3697	0.3868	0.2175	0.4118	0.9408	0.9626	0.9723	0.6622	0.8348	0.2343	0.386	0.6988	0.7632	0.118	0.2534	0.0435	0.1986	0.0054	0.0939	0.0	0.0	0.121	0.2462	0.0254	0.12
16.3350	5.0	775	25.6590	0.2832	0.4136	0.2814	0.1205	0.3136	0.9403	0.1807	0.3585	0.3736	0.2062	0.3962	0.9428	0.9703	0.9783	0.6595	0.8328	0.2347	0.3654	0.6683	0.7211	0.1216	0.2466	0.03	0.1736	0.0049	0.0909	0.0	0.0	0.1214	0.2474	0.0208	0.08
16.0730	6.0	930	25.5466	0.2887	0.4254	0.2901	0.1239	0.3183	0.9385	0.1857	0.37	0.3867	0.2171	0.4078	0.9416	0.9651	0.9747	0.6593	0.8308	0.2417	0.3832	0.7016	0.7684	0.1266	0.2638	0.0402	0.1833	0.0048	0.097	0.0006	0.025	0.126	0.241	0.0217	0.1
15.8823	7.0	1085	25.9349	0.2877	0.4215	0.2863	0.122	0.3234	0.9401	0.1823	0.3764	0.3956	0.2255	0.4378	0.9424	0.9698	0.9771	0.6594	0.8298	0.2411	0.3732	0.7049	0.7737	0.1227	0.2707	0.0414	0.2028	0.0051	0.0848	0.0008	0.05	0.1218	0.2538	0.01	0.14
15.5856	8.0	1240	25.6096	0.2891	0.4289	0.2905	0.1226	0.3211	0.9391	0.1913	0.379	0.3951	0.2312	0.4162	0.942	0.9667	0.9759	0.6605	0.8228	0.2371	0.3737	0.7092	0.7579	0.1303	0.2707	0.0372	0.2028	0.0049	0.1061	0.0021	0.1	0.1302	0.241	0.0129	0.1
15.2916	9.0	1395	25.5277	0.2904	0.4282	0.2877	0.1228	0.3212	0.9388	0.1935	0.3802	0.3982	0.2291	0.4252	0.9412	0.966	0.9735	0.6601	0.8248	0.2396	0.3883	0.7266	0.7737	0.1227	0.2759	0.0365	0.1931	0.0067	0.0939	0.0023	0.075	0.1333	0.2436	0.0102	0.14
15.1850	10.0	1550	26.6500	0.2861	0.425	0.2873	0.1204	0.3288	0.9391	0.1955	0.3752	0.3935	0.2283	0.4298	0.9416	0.9668	0.9747	0.659	0.8286	0.2331	0.3693	0.6927	0.7263	0.1201	0.2672	0.0486	0.2292	0.005	0.0758	0.0028	0.1	0.1192	0.2436	0.0135	0.12

Framework versions

—Transformers 5.0.0
—Pytorch 2.10.0+cu128
—Datasets 4.8.3
—Tokenizers 0.22.2