This model is a fine-tuned version of nvidia/segformer-b5-finetuned-ade-640-640 on the AllanK24/apple-dms-materials dataset. It achieves the following results on the evaluation set:

Loss: 1.2879
Mean Iou: 0.4724
Mean Accuracy: 0.5648
Overall Accuracy: 0.8240
Accuracy Animal Skin: 0.6669
Iou Animal Skin: 0.4954
Accuracy Bone Teeth Horn: 0.0000
Iou Bone Teeth Horn: 0.0000
Accuracy Brickwork: 0.7143
Iou Brickwork: 0.5850
Accuracy Cardboard: 0.6449
Iou Cardboard: 0.4995
Accuracy Carpet Rug: 0.8798
Iou Carpet Rug: 0.7598
Accuracy Ceiling Tile: 0.8870
Iou Ceiling Tile: 0.7860
Accuracy Ceramic: 0.7790
Iou Ceramic: 0.6400
Accuracy Chalkboard Blackboard: 0.7807
Iou Chalkboard Blackboard: 0.6577
Accuracy Clutter: 0.0578
Iou Clutter: 0.0491
Accuracy Concrete: 0.6090
Iou Concrete: 0.4245
Accuracy Cork Corkboard: 0.0904
Iou Cork Corkboard: 0.0887
Accuracy Engineered Stone: 0.1158
Iou Engineered Stone: 0.1017
Accuracy Fabric Cloth: 0.9022
Iou Fabric Cloth: 0.8034
Accuracy Fiberglass Wool: 0.0
Iou Fiberglass Wool: 0.0
Accuracy Fire: 0.5132
Iou Fire: 0.4591
Accuracy Foliage: 0.9381
Iou Foliage: 0.8476
Accuracy Food: 0.9138
Iou Food: 0.7933
Accuracy Fur: 0.9282
Iou Fur: 0.8474
Accuracy Gemstone Quartz: 0.5965
Iou Gemstone Quartz: 0.4286
Accuracy Glass: 0.7541
Iou Glass: 0.6325
Accuracy Hair: 0.8567
Iou Hair: 0.7472
Accuracy Ice: 0.2807
Iou Ice: 0.2218
Accuracy Leather: 0.6177
Iou Leather: 0.5098
Accuracy Liquid Non-water: 0.3093
Iou Liquid Non-water: 0.2761
Accuracy Metal: 0.4917
Iou Metal: 0.3705
Accuracy Mirror: 0.5999
Iou Mirror: 0.5131
Accuracy Paint Plaster Enamel: 0.8792
Iou Paint Plaster Enamel: 0.7555
Accuracy Paper: 0.7460
Iou Paper: 0.6039
Accuracy Pearl: 0.0
Iou Pearl: 0.0
Accuracy Photograph Painting: 0.4887
Iou Photograph Painting: 0.3704
Accuracy Plastic Clear: 0.3719
Iou Plastic Clear: 0.2842
Accuracy Plastic Non-clear: 0.5611
Iou Plastic Non-clear: 0.4098
Accuracy Rubber Latex: 0.3284
Iou Rubber Latex: 0.2964
Accuracy Sand: 0.6453
Iou Sand: 0.5073
Accuracy Skin Lips: 0.8633
Iou Skin Lips: 0.7585
Accuracy Sky: 0.9738
Iou Sky: 0.9361
Accuracy Snow: 0.6948
Iou Snow: 0.5942
Accuracy Soap: 0.0
Iou Soap: 0.0
Accuracy Soil Mud: 0.5642
Iou Soil Mud: 0.4117
Accuracy Sponge: 0.0
Iou Sponge: 0.0
Accuracy Stone Natural: 0.7217
Iou Stone Natural: 0.5622
Accuracy Stone Polished: 0.3487
Iou Stone Polished: 0.2630
Accuracy Styrofoam: 0.0
Iou Styrofoam: 0.0
Accuracy Tile: 0.8132
Iou Tile: 0.6839
Accuracy Wallpaper: 0.6192
Iou Wallpaper: 0.4637
Accuracy Water: 0.9112
Iou Water: 0.8091
Accuracy Wax: 0.5252
Iou Wax: 0.4924
Accuracy Whiteboard: 0.7797
Iou Whiteboard: 0.6631
Accuracy Wicker: 0.5724
Iou Wicker: 0.5007
Accuracy Wood: 0.8634
Iou Wood: 0.7459
Accuracy Wood Tree: 0.5530
Iou Wood Tree: 0.4346
Accuracy Asphalt: 0.6158
Iou Asphalt: 0.4780

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: 0.001
train_batch_size: 32
eval_batch_size: 16
seed: 42
distributed_type: multi-GPU
num_devices: 8
total_train_batch_size: 256
total_eval_batch_size: 128
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: cosine
lr_scheduler_warmup_steps: 0.1
num_epochs: 20
label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Animal Skin	Iou Animal Skin	Accuracy Bone Teeth Horn	Iou Bone Teeth Horn	Accuracy Brickwork	Iou Brickwork	Accuracy Cardboard	Iou Cardboard	Accuracy Carpet Rug	Iou Carpet Rug	Accuracy Ceiling Tile	Iou Ceiling Tile	Accuracy Ceramic	Iou Ceramic	Accuracy Chalkboard Blackboard	Iou Chalkboard Blackboard	Accuracy Clutter	Iou Clutter	Accuracy Concrete	Iou Concrete	Accuracy Cork Corkboard	Iou Cork Corkboard	Accuracy Engineered Stone	Iou Engineered Stone	Accuracy Fabric Cloth	Iou Fabric Cloth	Accuracy Fire	Iou Fire	Accuracy Foliage	Iou Foliage	Accuracy Food	Iou Food	Accuracy Fur	Iou Fur	Accuracy Gemstone Quartz	Iou Gemstone Quartz	Accuracy Glass	Iou Glass	Accuracy Hair	Iou Hair	Accuracy Ice	Iou Ice	Accuracy Leather	Iou Leather	Accuracy Liquid Non-water	Iou Liquid Non-water	Accuracy Metal	Iou Metal	Accuracy Mirror	Iou Mirror	Accuracy Paint Plaster Enamel	Iou Paint Plaster Enamel	Accuracy Paper	Iou Paper	Accuracy Photograph Painting	Iou Photograph Painting	Accuracy Plastic Clear	Iou Plastic Clear	Accuracy Plastic Non-clear	Iou Plastic Non-clear	Accuracy Rubber Latex	Iou Rubber Latex	Accuracy Sand	Iou Sand	Accuracy Skin Lips	Iou Skin Lips	Accuracy Sky	Iou Sky	Accuracy Snow	Iou Snow	Accuracy Soil Mud	Iou Soil Mud	Accuracy Stone Natural	Iou Stone Natural	Accuracy Stone Polished	Iou Stone Polished	Accuracy Tile	Iou Tile	Accuracy Wallpaper	Iou Wallpaper	Accuracy Water	Iou Water	Accuracy Wax	Iou Wax	Accuracy Whiteboard	Iou Whiteboard	Accuracy Wicker	Iou Wicker	Accuracy Wood	Iou Wood	Accuracy Wood Tree	Iou Wood Tree	Accuracy Asphalt	Iou Asphalt
2.3390	1.7045	150	1.3873	0.2906	0.3589	0.7696	0.0	0.0	0.0	0.0	0.5911	0.4554	0.2521	0.2423	0.8704	0.6817	0.8120	0.7332	0.6779	0.5571	0.0244	0.0244	0.0	0.0	0.5575	0.3485	0.0	0.0	0.0	0.0	0.9057	0.7270	0.0	0.0	0.9296	0.8088	0.8791	0.7569	0.9042	0.7140	0.0	0.0	0.6939	0.5522	0.7899	0.6600	0.0	0.0	0.2282	0.2050	0.0	0.0	0.3187	0.2535	0.4148	0.3683	0.8762	0.7060	0.7902	0.4873	0.4123	0.2827	0.0103	0.0102	0.3918	0.2991	0.0	0.0	0.0004	0.0004	0.7840	0.6628	0.9583	0.9131	0.0	0.0	0.5761	0.2977	0.6095	0.4344	0.0191	0.0191	0.7104	0.6076	0.2162	0.2018	0.9159	0.7915	0.0	0.0	0.3125	0.2751	0.0	0.0	0.8002	0.6859	0.0000	0.0000	0.4325	0.3506
1.2336	3.4091	300	1.2783	0.4077	0.5000	0.8047	0.4999	0.3737	0.0	0.0	0.7482	0.5652	0.5903	0.4623	0.9128	0.6888	0.8757	0.7657	0.7210	0.6021	0.7550	0.5898	0.0	0.0	0.6337	0.3852	0.0	0.0	0.0	0.0	0.9002	0.7811	0.0001	0.0001	0.9394	0.8322	0.9096	0.7764	0.9300	0.8306	0.0	0.0	0.6882	0.5918	0.8126	0.7093	0.0027	0.0027	0.5516	0.4738	0.0	0.0	0.4638	0.3340	0.5795	0.4795	0.8712	0.7364	0.7824	0.5690	0.5541	0.3514	0.3345	0.2470	0.4359	0.3490	0.2426	0.2211	0.7089	0.5633	0.8241	0.7167	0.9711	0.9269	0.6966	0.6169	0.5353	0.4179	0.7151	0.5136	0.2438	0.2084	0.6999	0.6283	0.6526	0.4858	0.9225	0.7855	0.0	0.0	0.7310	0.6201	0.5487	0.4628	0.8496	0.7238	0.4551	0.3419	0.7105	0.4702
1.0961	5.1136	450	1.2512	0.4312	0.5160	0.8156	0.5781	0.4693	0.0	0.0	0.7102	0.5866	0.7011	0.4708	0.8697	0.7551	0.8780	0.7657	0.7688	0.6118	0.7483	0.6313	0.0	0.0	0.5257	0.3795	0.0	0.0	0.0	0.0	0.8948	0.7899	0.3314	0.3072	0.9344	0.8414	0.9010	0.7806	0.9418	0.8521	0.0016	0.0016	0.7596	0.6203	0.8280	0.7256	0.3399	0.2757	0.5542	0.4792	0.0	0.0	0.4231	0.3395	0.6098	0.5003	0.8862	0.7450	0.7694	0.5842	0.4859	0.3539	0.3158	0.2584	0.5146	0.3852	0.2588	0.2400	0.6976	0.5494	0.8379	0.7323	0.9751	0.9361	0.7754	0.6701	0.6294	0.4429	0.6618	0.5499	0.3015	0.2526	0.8294	0.6686	0.5842	0.4540	0.9128	0.8011	0.0	0.0	0.7076	0.6225	0.5282	0.4530	0.8322	0.7307	0.4905	0.3869	0.5363	0.4241
1.0257	6.8182	600	1.2558	0.4346	0.5253	0.8164	0.6225	0.4611	0.0	0.0	0.7154	0.5722	0.6348	0.4974	0.8897	0.7510	0.8634	0.7681	0.7403	0.6255	0.7590	0.6489	0.0	0.0	0.6283	0.4189	0.0	0.0	0.0000	0.0000	0.9042	0.7936	0.5297	0.4131	0.9337	0.8454	0.8899	0.7800	0.9302	0.8440	0.0219	0.0199	0.7718	0.6245	0.8424	0.7336	0.2455	0.2031	0.5605	0.4826	0.0004	0.0004	0.5314	0.3672	0.5960	0.5065	0.8611	0.7481	0.7367	0.5936	0.4883	0.3499	0.3148	0.2629	0.5460	0.3996	0.2981	0.2670	0.6377	0.5024	0.8457	0.7415	0.9740	0.9371	0.6959	0.6111	0.5811	0.4149	0.7289	0.5593	0.4109	0.2372	0.7779	0.6653	0.6314	0.4681	0.9029	0.7937	0.0	0.0	0.7692	0.6271	0.5367	0.4704	0.8604	0.7387	0.4847	0.4003	0.6200	0.4518
0.9864	8.5227	750	1.2668	0.4451	0.5371	0.8195	0.6404	0.4525	0.0	0.0	0.7251	0.5979	0.6371	0.4917	0.8974	0.7429	0.8887	0.7706	0.7607	0.6339	0.7797	0.6545	0.0	0.0	0.5719	0.4187	0.0	0.0	0.0741	0.0717	0.9051	0.7970	0.4362	0.3608	0.9378	0.8420	0.8936	0.7763	0.9328	0.8408	0.3477	0.2751	0.7471	0.6271	0.8404	0.7371	0.3318	0.2704	0.5480	0.4779	0.0000	0.0000	0.4938	0.3654	0.6066	0.5084	0.8625	0.7512	0.7469	0.5973	0.5093	0.3695	0.3599	0.2762	0.5718	0.4083	0.3096	0.2828	0.5965	0.4762	0.8525	0.7473	0.9735	0.9388	0.7473	0.6344	0.5148	0.3931	0.7570	0.5384	0.3941	0.2692	0.8032	0.6809	0.6762	0.4575	0.9033	0.8029	0.0145	0.0145	0.7590	0.6708	0.5315	0.4777	0.8777	0.7409	0.5186	0.4222	0.6536	0.4814
0.9613	10.2273	900	1.2749	0.4598	0.5506	0.8220	0.6409	0.4965	0.0	0.0	0.7162	0.5950	0.6576	0.5049	0.8908	0.7572	0.8743	0.7732	0.7820	0.6344	0.7747	0.6545	0.0099	0.0096	0.6239	0.4197	0.0001	0.0001	0.0686	0.0653	0.8969	0.8006	0.6038	0.5146	0.9392	0.8437	0.9172	0.7885	0.9292	0.8466	0.4658	0.3577	0.7317	0.6235	0.8540	0.7404	0.3206	0.2571	0.6258	0.5084	0.0893	0.0856	0.4680	0.3602	0.5891	0.5095	0.8830	0.7530	0.7496	0.6016	0.5025	0.3692	0.3680	0.2810	0.5549	0.4037	0.3307	0.2918	0.6214	0.4954	0.8673	0.7511	0.9746	0.9346	0.7524	0.6199	0.5928	0.4194	0.7180	0.5696	0.3411	0.2665	0.7995	0.6859	0.5284	0.4289	0.8983	0.8131	0.2922	0.2912	0.7520	0.6422	0.5767	0.4885	0.8678	0.7430	0.5317	0.4240	0.6595	0.4911
0.9432	11.9318	1050	1.2817	0.4623	0.5496	0.8226	0.6260	0.4877	0.0	0.0	0.6989	0.5877	0.6410	0.4947	0.8766	0.7568	0.8740	0.7743	0.7771	0.6412	0.7750	0.6609	0.0248	0.0232	0.6046	0.4193	0.0585	0.0571	0.1144	0.1048	0.9023	0.8013	0.5393	0.4625	0.9421	0.8467	0.9153	0.7873	0.9294	0.8453	0.4383	0.3306	0.7254	0.6232	0.8586	0.7431	0.3814	0.2985	0.6159	0.5046	0.0755	0.0739	0.4769	0.3670	0.5529	0.4904	0.8873	0.7540	0.7449	0.6012	0.4669	0.3610	0.3741	0.2872	0.5452	0.4047	0.3194	0.2856	0.6295	0.4897	0.8586	0.7535	0.9750	0.9361	0.7493	0.6305	0.5527	0.4120	0.7287	0.5613	0.3395	0.2515	0.8147	0.6862	0.5239	0.4191	0.8912	0.8066	0.3966	0.3871	0.8114	0.6855	0.5755	0.4976	0.8694	0.7423	0.5092	0.4187	0.5901	0.4840
0.9319	13.6364	1200	1.2814	0.4681	0.5584	0.8229	0.6690	0.4890	0.0	0.0	0.7132	0.5866	0.6587	0.5031	0.8829	0.7588	0.8878	0.7851	0.7761	0.6344	0.7906	0.6609	0.0456	0.0404	0.6174	0.4232	0.0963	0.0924	0.0978	0.0888	0.9000	0.8026	0.5484	0.4819	0.9412	0.8476	0.9087	0.7878	0.9226	0.8475	0.3579	0.2824	0.7477	0.6297	0.8581	0.7449	0.2774	0.2280	0.5790	0.4931	0.2721	0.2474	0.4840	0.3674	0.6042	0.5082	0.8790	0.7548	0.7380	0.6034	0.4859	0.3678	0.3655	0.2831	0.5791	0.4102	0.3235	0.2871	0.6424	0.5080	0.8606	0.7559	0.9728	0.9347	0.6963	0.5990	0.5599	0.4132	0.7145	0.5539	0.3639	0.2720	0.8225	0.6854	0.5914	0.4453	0.9113	0.8092	0.5005	0.4807	0.8112	0.6960	0.5693	0.4964	0.8562	0.7430	0.5369	0.4303	0.6180	0.4825
0.9231	15.3409	1350	1.2847	0.4721	0.5654	0.8234	0.6722	0.4974	0.0	0.0	0.7089	0.5863	0.6577	0.5055	0.8780	0.7584	0.8826	0.7864	0.7689	0.6361	0.7811	0.6562	0.0552	0.0474	0.6089	0.4245	0.0915	0.0868	0.1090	0.0955	0.9012	0.8033	0.5363	0.4697	0.9417	0.8465	0.9120	0.7872	0.9246	0.8477	0.6218	0.4352	0.7524	0.6330	0.8560	0.7464	0.2909	0.2245	0.6221	0.5120	0.2903	0.2622	0.5041	0.3740	0.6073	0.5184	0.8750	0.7543	0.7469	0.6048	0.4879	0.3687	0.3642	0.2825	0.5704	0.4110	0.3203	0.2892	0.6391	0.5055	0.8644	0.7575	0.9746	0.9366	0.7025	0.5964	0.5722	0.4120	0.7218	0.5555	0.3704	0.2687	0.8160	0.6859	0.6180	0.4596	0.9101	0.8122	0.5197	0.4915	0.7785	0.6662	0.5701	0.5001	0.8630	0.7436	0.5315	0.4309	0.6109	0.4783
0.9183	17.0455	1500	1.2863	0.4713	0.5631	0.8237	0.6723	0.4992	0.0000	0.0000	0.7099	0.5837	0.6476	0.5016	0.8786	0.7602	0.8819	0.7854	0.7827	0.6402	0.7817	0.6586	0.0575	0.0494	0.6065	0.4214	0.0870	0.0841	0.0944	0.0861	0.8997	0.8043	0.5037	0.4502	0.9389	0.8479	0.9153	0.7933	0.9268	0.8484	0.5980	0.4333	0.7499	0.6317	0.8551	0.7469	0.2734	0.2177	0.6105	0.5083	0.3039	0.2723	0.4996	0.3718	0.5976	0.5147	0.8793	0.7553	0.7483	0.6025	0.4879	0.3691	0.3695	0.2849	0.5659	0.4103	0.3233	0.2930	0.6398	0.5036	0.8637	0.7585	0.9738	0.9361	0.6863	0.5927	0.5698	0.4108	0.7246	0.5633	0.3730	0.2655	0.8143	0.6845	0.5799	0.4443	0.9120	0.8108	0.5231	0.4917	0.7773	0.6611	0.5646	0.4991	0.8638	0.7457	0.5475	0.4334	0.6215	0.4824
0.9157	18.75	1650	1.2879	0.4724	0.5648	0.8240	0.6669	0.4954	0.0000	0.0000	0.7143	0.5850	0.6449	0.4995	0.8798	0.7598	0.8870	0.7860	0.7790	0.6400	0.7807	0.6577	0.0578	0.0491	0.6090	0.4245	0.0904	0.0887	0.1158	0.1017	0.9022	0.8034	0.5132	0.4591	0.9381	0.8476	0.9138	0.7933	0.9282	0.8474	0.5965	0.4286	0.7541	0.6325	0.8567	0.7472	0.2807	0.2218	0.6177	0.5098	0.3093	0.2761	0.4917	0.3705	0.5999	0.5131	0.8792	0.7555	0.7460	0.6039	0.4887	0.3704	0.3719	0.2842	0.5611	0.4098	0.3284	0.2964	0.6453	0.5073	0.8633	0.7585	0.9738	0.9361	0.6948	0.5942	0.5642	0.4117	0.7217	0.5622	0.3487	0.2630	0.8132	0.6839	0.6192	0.4637	0.9112	0.8091	0.5252	0.4924	0.7797	0.6631	0.5724	0.5007	0.8634	0.7459	0.5530	0.4346	0.6158	0.4780

Framework versions

Transformers 5.0.0
Pytorch 2.9.1+cu128
Datasets 4.5.0
Tokenizers 0.22.2

AllanK24/segformer-b5-finetuned-apple-dms-run9

segformer-b5-finetuned-apple-dms-run9

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

AllanK24/
segformer-b5-finetuned-apple-dms-run9