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: 13.7972
Mean Iou: 0.1465
Mean Accuracy: 0.1831
Overall Accuracy: 0.6387
Iou No Label: 0.0
Accuracy Animal Skin: 0.0
Iou Animal Skin: 0.0
Accuracy Bone Teeth Horn: 0.0
Iou Bone Teeth Horn: 0.0
Accuracy Brickwork: 0.0
Iou Brickwork: 0.0
Accuracy Cardboard: 0.0
Iou Cardboard: 0.0
Accuracy Carpet Rug: 0.7365
Iou Carpet Rug: 0.6151
Accuracy Ceiling Tile: 0.7464
Iou Ceiling Tile: 0.6825
Accuracy Ceramic: 0.0
Iou Ceramic: 0.0
Accuracy Chalkboard Blackboard: 0.0
Iou Chalkboard Blackboard: 0.0
Accuracy Clutter: 0.0
Iou Clutter: 0.0
Accuracy Concrete: 0.0
Iou Concrete: 0.0
Accuracy Cork Corkboard: 0.0
Iou Cork Corkboard: 0.0
Accuracy Engineered Stone: 0.0
Iou Engineered Stone: 0.0
Accuracy Fabric Cloth: 0.8713
Iou Fabric Cloth: 0.6744
Accuracy Fiberglass Wool: 0.0
Iou Fiberglass Wool: 0.0
Accuracy Fire: 0.0
Iou Fire: 0.0
Accuracy Foliage: 0.9291
Iou Foliage: 0.5950
Accuracy Food: 0.4894
Iou Food: 0.4539
Accuracy Fur: 0.0000
Iou Fur: 0.0000
Accuracy Gemstone Quartz: 0.0
Iou Gemstone Quartz: 0.0
Accuracy Glass: 0.6571
Iou Glass: 0.4831
Accuracy Hair: 0.1408
Iou Hair: 0.1379
Accuracy Ice: 0.0
Iou Ice: 0.0
Accuracy Leather: 0.0
Iou Leather: 0.0
Accuracy Liquid Non-water: 0.0
Iou Liquid Non-water: 0.0
Accuracy Metal: 0.0
Iou Metal: 0.0
Accuracy Mirror: 0.0
Iou Mirror: 0.0
Accuracy Paint Plaster Enamel: 0.7900
Iou Paint Plaster Enamel: 0.6728
Accuracy Paper: 0.3281
Iou Paper: 0.2929
Accuracy Pearl: 0.0
Iou Pearl: 0.0
Accuracy Photograph Painting: 0.0
Iou Photograph Painting: 0.0
Accuracy Plastic Clear: 0.0
Iou Plastic Clear: 0.0
Accuracy Plastic Non-clear: 0.0003
Iou Plastic Non-clear: 0.0003
Accuracy Rubber Latex: 0.0
Iou Rubber Latex: 0.0
Accuracy Sand: 0.0
Iou Sand: 0.0
Accuracy Skin Lips: 0.5098
Iou Skin Lips: 0.3949
Accuracy Sky: 0.9492
Iou Sky: 0.8934
Accuracy Snow: 0.0
Iou Snow: 0.0
Accuracy Soap: 0.0
Iou Soap: 0.0
Accuracy Soil Mud: 0.0
Iou Soil Mud: 0.0
Accuracy Sponge: 0.0
Iou Sponge: 0.0
Accuracy Stone Natural: 0.0119
Iou Stone Natural: 0.0117
Accuracy Stone Polished: 0.0
Iou Stone Polished: 0.0
Accuracy Styrofoam: 0.0
Iou Styrofoam: 0.0
Accuracy Tile: 0.7035
Iou Tile: 0.5573
Accuracy Wallpaper: 0.0
Iou Wallpaper: 0.0
Accuracy Water: 0.8647
Iou Water: 0.6487
Accuracy Wax: 0.0
Iou Wax: 0.0
Accuracy Whiteboard: 0.0
Iou Whiteboard: 0.0
Accuracy Wicker: 0.0
Iou Wicker: 0.0
Accuracy Wood: 0.7930
Iou Wood: 0.6508
Accuracy Wood Tree: 0.0
Iou Wood Tree: 0.0
Accuracy Asphalt: 0.0
Iou Asphalt: 0.0

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
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: linear
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Cardboard	Iou Cardboard	Accuracy Carpet Rug	Iou Carpet Rug	Accuracy Ceiling Tile	Iou Ceiling Tile	Accuracy Concrete	Iou Concrete	Accuracy Fabric Cloth	Iou Fabric Cloth	Accuracy Foliage	Iou Foliage	Accuracy Food	Iou Food	Accuracy Fur	Iou Fur	Accuracy Glass	Iou Glass	Accuracy Hair	Iou Hair	Iou Liquid Non-water	Accuracy Metal	Iou Metal	Iou Mirror	Iou Not On List	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 Sand	Iou Sand	Accuracy Skin Lips	Iou Skin Lips	Accuracy Sky	Iou Sky	Accuracy Sponge	Iou Sponge	Accuracy Stone Natural	Iou Stone Natural	Accuracy Styrofoam	Iou Styrofoam	Accuracy Tile	Iou Tile	Accuracy Wallpaper	Iou Wallpaper	Accuracy Water	Iou Water	Accuracy Wicker	Iou Wicker	Accuracy Wood	Iou Wood	Iou Bad Polygon	Accuracy Asphalt	Iou Asphalt
27.5939	2.2727	200	23.5569	0.0619	0.0994	0.5271	0.0000	0.0000	0.0210	0.0208	0.0349	0.0348	0.0000	0.0000	0.8433	0.5460	0.8881	0.3613	0.0002	0.0002	0.0	0.0	0.3023	0.2407	0.0015	0.0014	0.0	0.0000	0.0000	0.0	0.0	0.7360	0.6027	0.0002	0.0002	0.0000	0.0000	0.0	0.0	0.0103	0.0097	0.0	0.0	0.0009	0.0009	0.9448	0.8326	0.0	0.0	0.0000	0.0000	0.0	0.0	0.5905	0.3089	0.0	0.0	0.0083	0.0083	0.0	0.0	0.7857	0.4981	0.0	0.0	0.0
21.6200	4.5455	400	19.7415	0.0915	0.1295	0.5936	0.0	0.0	0.3420	0.3187	0.3897	0.3806	0.0	0.0	0.8619	0.5887	0.9098	0.4627	0.0046	0.0046	0.0	0.0	0.4988	0.3468	0.0048	0.0047	0.0	0.0	0.0	0.8109	0.6419	0.0001	0.0001	0.0	0.0	0.0	0.0	0.0009	0.0009	0.0	0.0	0.0060	0.0060	0.9431	0.8686	0.0	0.0	0.0	0.0	0.0	0.0	0.7232	0.4043	0.0	0.0	0.4001	0.3559	0.0	0.0	0.8360	0.5561	0.0	0.0	0.0
18.5498	6.8182	600	17.4553	0.1118	0.1488	0.6158	0.0	0.0	0.6324	0.5242	0.5825	0.5523	0.0	0.0	0.8647	0.6069	0.9223	0.5422	0.0511	0.0507	0.0	0.0	0.5855	0.3873	0.0124	0.0123	0.0000	0.0000	0.0	0.8118	0.6581	0.0002	0.0002	0.0	0.0	0.0	0.0	0.0002	0.0002	0.0	0.0	0.0461	0.0451	0.9481	0.8765	0.0	0.0	0.0001	0.0001	0.0	0.0	0.7229	0.4941	0.0	0.0	0.7340	0.5760	0.0	0.0	0.8224	0.5998	0.0	0.0
16.5349	9.0909	800	16.0506	0.1202	0.1574	0.6225	0.0	0.0	0.7067	0.5744	0.6436	0.6048	0.0	0.0	0.8733	0.6141	0.9253	0.5585	0.1829	0.1802	0.0	0.0	0.6468	0.4215	0.0085	0.0085	0.0	0.0	0.0	0.8026	0.6645	0.0065	0.0065	0.0	0.0	0.0	0.0	0.0002	0.0002	0.0	0.0	0.0953	0.0902	0.9415	0.8810	0.0	0.0	0.0033	0.0033	0.0	0.0	0.7327	0.5148	0.0	0.0	0.8148	0.6209	0.0	0.0	0.8022	0.6287	0.0	0.0
15.1469	11.3636	1000	14.9962	0.1300	0.1665	0.6314	0.0	0.0	0.7266	0.5961	0.6857	0.6381	0.0	0.0	0.8737	0.6373	0.9264	0.5831	0.3187	0.3089	0.0	0.0	0.6238	0.4449	0.0179	0.0177	0.0	0.0	0.0	0.8120	0.6690	0.0806	0.0789	0.0	0.0	0.0	0.0	0.0003	0.0003	0.0	0.0	0.2666	0.2299	0.9436	0.8842	0.0	0.0	0.0129	0.0127	0.0	0.0	0.7142	0.5329	0.0	0.0	0.8486	0.6230	0.0	0.0	0.8078	0.6325	0.0	0.0
14.2406	13.6364	1200	14.3933	0.1383	0.1752	0.6352	0.0	0.0	0.7366	0.6052	0.7381	0.6737	0.0	0.0	0.8767	0.6493	0.9291	0.5855	0.4061	0.3868	0.0	0.0	0.6420	0.4662	0.0493	0.0489	0.0	0.0	0.0	0.7951	0.6715	0.2065	0.1953	0.0	0.0	0.0	0.0	0.0003	0.0003	0.0	0.0	0.3797	0.3106	0.9508	0.8907	0.0	0.0	0.0114	0.0113	0.0	0.0	0.7298	0.5449	0.0	0.0	0.8553	0.6478	0.0	0.0	0.8030	0.6422	0.0	0.0
13.6063	15.9091	1400	13.9943	0.1439	0.1809	0.6375	0.0	0.0	0.7382	0.6117	0.7529	0.6831	0.0	0.0	0.8777	0.6614	0.9289	0.5959	0.4785	0.4456	0.0	0.0	0.6431	0.4791	0.0968	0.0953	0.0	0.0	0.0	0.7890	0.6717	0.2932	0.2664	0.0	0.0	0.0	0.0	0.0003	0.0003	0.0	0.0	0.4660	0.3657	0.9505	0.8925	0.0	0.0	0.0151	0.0149	0.0	0.0	0.7125	0.5531	0.0	0.0	0.8667	0.6407	0.0	0.0	0.7993	0.6474	0.0	0.0
13.2493	18.1818	1600	13.7972	0.1465	0.1831	0.6387	0.0	0.0	0.7365	0.6151	0.7464	0.6825	0.0	0.0	0.8713	0.6744	0.9291	0.5950	0.4894	0.4539	0.0000	0.0000	0.6571	0.4831	0.1408	0.1379	0.0	0.0	0.0	0.7900	0.6728	0.3281	0.2929	0.0	0.0	0.0	0.0	0.0003	0.0003	0.0	0.0	0.5098	0.3949	0.9492	0.8934	0.0	0.0	0.0119	0.0117	0.0	0.0	0.7035	0.5573	0.0	0.0	0.8647	0.6487	0.0	0.0	0.7930	0.6508	0.0	0.0

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-run1

segformer-b5-finetuned-apple-dms-run1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

AllanK24/
segformer-b5-finetuned-apple-dms-run1