segformer-b0-finetuned-segments-SixrayGun8-14-2024

This model is a fine-tuned version of nvidia/mit-b0 on the saad7489/SIXray_Gun dataset. It achieves the following results on the evaluation set:

• Loss: 0.1155
• Mean Iou: 0.1748
• Mean Accuracy: 0.2486
• Overall Accuracy: 0.5704
• Accuracy No-label: nan
• Accuracy Object1: 0.6638
• Accuracy Object2: 0.5791
• Accuracy Object3: 0.0
• Accuracy Object4: 0.0
• Accuracy Object5: 0.0
• Accuracy Object6: nan
• Iou No-label: 0.0
• Iou Object1: 0.5674
• Iou Object2: 0.4817
• Iou Object3: 0.0
• Iou Object4: 0.0
• Iou Object5: 0.0
• Iou Object6: nan

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: 6e-05
• train_batch_size: 20
• eval_batch_size: 20
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy No-label	Accuracy Object1	Accuracy Object2	Accuracy Object3	Accuracy Object4	Accuracy Object5	Accuracy Object6	Iou No-label	Iou Object1	Iou Object2	Iou Object3	Iou Object4	Iou Object5	Iou Object6
1.5955	0.5128	20	1.7043	0.0922	0.1760	0.4766	nan	0.6799	0.1943	0.0058	0.0	0.0	nan	0.0	0.4999	0.1400	0.0058	0.0	0.0	0.0
1.2095	1.0256	40	1.2385	0.0865	0.1398	0.4206	nan	0.6565	0.0425	0.0	0.0	0.0	nan	0.0	0.4787	0.0403	0.0	0.0	0.0	nan
1.0262	1.5385	60	1.0043	0.0795	0.1256	0.3824	nan	0.6026	0.0256	0.0	0.0	0.0	nan	0.0	0.4521	0.0250	0.0	0.0	0.0	nan
0.935	2.0513	80	0.8920	0.0771	0.1201	0.3655	nan	0.5759	0.0245	0.0	0.0	0.0	nan	0.0	0.4386	0.0240	0.0	0.0	0.0	nan
0.8168	2.5641	100	0.8136	0.0691	0.1079	0.3359	nan	0.5386	0.0011	0.0	0.0	0.0	nan	0.0	0.4137	0.0011	0.0	0.0	0.0	nan
0.721	3.0769	120	0.7429	0.0758	0.1222	0.3803	nan	0.6100	0.0009	0.0	0.0	0.0	nan	0.0	0.4538	0.0009	0.0	0.0	0.0	nan
0.6536	3.5897	140	0.6914	0.0828	0.1378	0.4292	nan	0.6886	0.0003	0.0	0.0	0.0	nan	0.0	0.4965	0.0003	0.0	0.0	0.0	nan
0.6535	4.1026	160	0.5916	0.0691	0.1091	0.3398	nan	0.5454	0.0000	0.0	0.0	0.0	nan	0.0	0.4147	0.0000	0.0	0.0	0.0	nan
0.5396	4.6154	180	0.5436	0.0780	0.1283	0.3996	nan	0.6409	0.0007	0.0	0.0	0.0	nan	0.0	0.4674	0.0007	0.0	0.0	0.0	nan
0.4738	5.1282	200	0.4589	0.0777	0.1275	0.3973	nan	0.6376	0.0000	0.0	0.0	0.0	nan	0.0	0.4664	0.0000	0.0	0.0	0.0	nan
0.4548	5.6410	220	0.3964	0.0739	0.1182	0.3679	nan	0.5902	0.0007	0.0	0.0	0.0	nan	0.0	0.4426	0.0007	0.0	0.0	0.0	nan
0.3918	6.1538	240	0.4124	0.0895	0.1507	0.4629	nan	0.7348	0.0185	0.0	0.0	0.0	nan	0.0	0.5187	0.0183	0.0	0.0	0.0	nan
0.3569	6.6667	260	0.3519	0.0859	0.1465	0.4560	nan	0.7314	0.0010	0.0	0.0	0.0	nan	0.0	0.5144	0.0010	0.0	0.0	0.0	nan
0.3143	7.1795	280	0.3189	0.0860	0.1464	0.4562	nan	0.7322	0.0	0.0	0.0	0.0	nan	0.0	0.5158	0.0	0.0	0.0	0.0	nan
0.3131	7.6923	300	0.3064	0.0940	0.1601	0.4880	nan	0.7698	0.0306	0.0	0.0	0.0	nan	0.0	0.5339	0.0302	0.0	0.0	0.0	nan
0.2623	8.2051	320	0.2687	0.0827	0.1392	0.4337	nan	0.6960	0.0001	0.0	0.0	0.0	nan	0.0	0.4961	0.0001	0.0	0.0	0.0	nan
0.2205	8.7179	340	0.2293	0.0838	0.1377	0.4232	nan	0.6722	0.0162	0.0	0.0	0.0	nan	0.0	0.4865	0.0161	0.0	0.0	0.0	nan
0.2316	9.2308	360	0.2277	0.0865	0.1445	0.4444	nan	0.7062	0.0160	0.0	0.0	0.0	nan	0.0	0.5032	0.0160	0.0	0.0	0.0	nan
0.2626	9.7436	380	0.2061	0.0936	0.1452	0.4223	nan	0.6405	0.0856	0.0	0.0	0.0	nan	0.0	0.4777	0.0836	0.0	0.0	0.0	nan
0.1835	10.2564	400	0.1938	0.1100	0.1683	0.4696	nan	0.6863	0.1551	0.0	0.0	0.0	nan	0.0	0.5108	0.1491	0.0	0.0	0.0	nan
0.2101	10.7692	420	0.1763	0.1127	0.1703	0.4652	nan	0.6659	0.1858	0.0	0.0	0.0	nan	0.0	0.5027	0.1732	0.0	0.0	0.0	nan
0.173	11.2821	440	0.1608	0.1284	0.1867	0.4838	nan	0.6555	0.2783	0.0	0.0	0.0	nan	0.0	0.5130	0.2576	0.0	0.0	0.0	nan
0.1614	11.7949	460	0.1650	0.1464	0.2140	0.5146	nan	0.6384	0.4316	0.0	0.0	0.0	nan	0.0	0.5170	0.3616	0.0	0.0	0.0	nan
0.1575	12.3077	480	0.1562	0.1489	0.2184	0.5282	nan	0.6601	0.4318	0.0	0.0	0.0	nan	0.0	0.5311	0.3622	0.0	0.0	0.0	nan
0.1681	12.8205	500	0.1562	0.1657	0.2429	0.5762	nan	0.7023	0.5121	0.0	0.0	0.0	nan	0.0	0.5699	0.4241	0.0	0.0	0.0	nan
0.1325	13.3333	520	0.1430	0.1585	0.2268	0.5429	nan	0.6693	0.4649	0.0	0.0	0.0	nan	0.0	0.5489	0.4024	0.0	0.0	0.0	nan
0.1414	13.8462	540	0.1283	0.1519	0.2156	0.5188	nan	0.6441	0.4336	0.0	0.0	0.0	nan	0.0	0.5311	0.3801	0.0	0.0	0.0	nan
0.1276	14.3590	560	0.1296	0.1399	0.1953	0.4706	nan	0.5855	0.3909	0.0	0.0	0.0	nan	0.0	0.4895	0.3501	0.0	0.0	0.0	nan
0.1384	14.8718	580	0.1291	0.1526	0.2160	0.5212	nan	0.6493	0.4307	0.0	0.0	0.0	nan	0.0	0.5334	0.3819	0.0	0.0	0.0	nan
0.1546	15.3846	600	0.1294	0.1692	0.2438	0.5627	nan	0.6604	0.5587	0.0	0.0	0.0	nan	0.0	0.5584	0.4571	0.0	0.0	0.0	nan
0.1308	15.8974	620	0.1219	0.1625	0.2305	0.5490	nan	0.6725	0.4798	0.0	0.0	0.0	nan	0.0	0.5564	0.4189	0.0	0.0	0.0	nan
0.1365	16.4103	640	0.1241	0.1686	0.2469	0.5603	nan	0.6418	0.5925	0.0	0.0	0.0	nan	0.0	0.5478	0.4640	0.0	0.0	0.0	nan
0.1361	16.9231	660	0.1191	0.1590	0.2255	0.5385	nan	0.6620	0.4653	0.0	0.0	0.0	nan	0.0	0.5464	0.4074	0.0	0.0	0.0	nan
0.1125	17.4359	680	0.1168	0.1693	0.2397	0.5549	nan	0.6537	0.5451	0.0	0.0	0.0	nan	0.0	0.5570	0.4589	0.0	0.0	0.0	nan
0.1134	17.9487	700	0.1182	0.1732	0.2470	0.5725	nan	0.6758	0.5592	0.0	0.0	0.0	nan	0.0	0.5705	0.4689	0.0	0.0	0.0	nan
0.1351	18.4615	720	0.1168	0.1751	0.2496	0.5698	nan	0.6580	0.5901	0.0	0.0	0.0	nan	0.0	0.5648	0.4859	0.0	0.0	0.0	nan
0.1382	18.9744	740	0.1177	0.1771	0.2530	0.5784	nan	0.6694	0.5956	0.0	0.0	0.0	nan	0.0	0.5724	0.4904	0.0	0.0	0.0	nan
0.1113	19.4872	760	0.1164	0.1789	0.2562	0.5859	nan	0.6784	0.6026	0.0	0.0	0.0	nan	0.0	0.5791	0.4943	0.0	0.0	0.0	nan
0.098	20.0	780	0.1155	0.1748	0.2486	0.5704	nan	0.6638	0.5791	0.0	0.0	0.0	nan	0.0	0.5674	0.4817	0.0	0.0	0.0	nan

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1