README.md

metadata

license: other
base_model: nvidia/mit-b0
tags:
  - vision
  - image-segmentation
  - generated_from_trainer
model-index:
  - name: segformer-b0-finetuned-segments-sidewalk-test
    results: []

segformer-b0-finetuned-segments-sidewalk-test

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

• Loss: 0.1893
• Mean Iou: 0.4552
• Mean Accuracy: 0.9104
• Overall Accuracy: 0.9104
• Accuracy Other: nan
• Accuracy Flat-sidewalk: 0.9104
• Iou Other: 0.0
• Iou Flat-sidewalk: 0.9104

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: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Other	Accuracy Flat-sidewalk	Iou Other	Iou Flat-sidewalk
0.4805	0.05	20	0.5080	0.4534	0.9069	0.9069	nan	0.9069	0.0	0.9069
0.2146	0.1	40	0.3937	0.4660	0.9319	0.9319	nan	0.9319	0.0	0.9319
0.215	0.15	60	0.3593	0.4476	0.8952	0.8952	nan	0.8952	0.0	0.8952
0.151	0.2	80	0.2834	0.4423	0.8845	0.8845	nan	0.8845	0.0	0.8845
0.174	0.25	100	0.3268	0.4612	0.9225	0.9225	nan	0.9225	0.0	0.9225
0.1597	0.3	120	0.2900	0.4229	0.8457	0.8457	nan	0.8457	0.0	0.8457
0.2165	0.35	140	0.2723	0.4411	0.8822	0.8822	nan	0.8822	0.0	0.8822
0.2	0.4	160	0.2598	0.4167	0.8334	0.8334	nan	0.8334	0.0	0.8334
0.577	0.45	180	0.3185	0.4708	0.9416	0.9416	nan	0.9416	0.0	0.9416
0.2466	0.5	200	0.2305	0.4295	0.8589	0.8589	nan	0.8589	0.0	0.8589
0.1742	0.55	220	0.2439	0.4544	0.9089	0.9089	nan	0.9089	0.0	0.9089
0.1764	0.6	240	0.2318	0.4359	0.8719	0.8719	nan	0.8719	0.0	0.8719
0.1432	0.65	260	0.2253	0.4318	0.8636	0.8636	nan	0.8636	0.0	0.8636
0.1472	0.7	280	0.2193	0.4353	0.8707	0.8707	nan	0.8707	0.0	0.8707
0.4737	0.75	300	0.2347	0.4407	0.8813	0.8813	nan	0.8813	0.0	0.8813
0.1567	0.8	320	0.2212	0.4248	0.8496	0.8496	nan	0.8496	0.0	0.8496
0.0832	0.85	340	0.2170	0.4426	0.8852	0.8852	nan	0.8852	0.0	0.8852
0.1718	0.9	360	0.2079	0.4390	0.8780	0.8780	nan	0.8780	0.0	0.8780
0.3256	0.95	380	0.2127	0.4576	0.9151	0.9151	nan	0.9151	0.0	0.9151
0.089	1.0	400	0.2249	0.4603	0.9207	0.9207	nan	0.9207	0.0	0.9207
0.103	1.05	420	0.2051	0.4360	0.8720	0.8720	nan	0.8720	0.0	0.8720
0.3474	1.1	440	0.2216	0.4333	0.8666	0.8666	nan	0.8666	0.0	0.8666
0.0851	1.15	460	0.2306	0.4681	0.9361	0.9361	nan	0.9361	0.0	0.9361
0.1989	1.2	480	0.2029	0.4516	0.9032	0.9032	nan	0.9032	0.0	0.9032
0.2072	1.25	500	0.2076	0.4666	0.9331	0.9331	nan	0.9331	0.0	0.9331
0.2898	1.3	520	0.2164	0.4645	0.9291	0.9291	nan	0.9291	0.0	0.9291
0.1578	1.35	540	0.2057	0.4457	0.8914	0.8914	nan	0.8914	0.0	0.8914
0.2697	1.4	560	0.1973	0.4646	0.9292	0.9292	nan	0.9292	0.0	0.9292
0.1269	1.45	580	0.1830	0.4467	0.8934	0.8934	nan	0.8934	0.0	0.8934
0.0908	1.5	600	0.1866	0.4471	0.8941	0.8941	nan	0.8941	0.0	0.8941
0.0614	1.55	620	0.1983	0.4632	0.9264	0.9264	nan	0.9264	0.0	0.9264
0.1043	1.6	640	0.1941	0.4598	0.9196	0.9196	nan	0.9196	0.0	0.9196
0.0532	1.65	660	0.1920	0.4553	0.9106	0.9106	nan	0.9106	0.0	0.9106
0.5912	1.7	680	0.1880	0.4530	0.9059	0.9059	nan	0.9059	0.0	0.9059
0.0604	1.75	700	0.1964	0.4611	0.9221	0.9221	nan	0.9221	0.0	0.9221
0.0899	1.8	720	0.1975	0.4623	0.9245	0.9245	nan	0.9245	0.0	0.9245
0.1153	1.85	740	0.1866	0.4580	0.9160	0.9160	nan	0.9160	0.0	0.9160
0.1038	1.9	760	0.1998	0.4652	0.9304	0.9304	nan	0.9304	0.0	0.9304
0.1448	1.95	780	0.1977	0.4624	0.9248	0.9248	nan	0.9248	0.0	0.9248
0.1298	2.0	800	0.1893	0.4552	0.9104	0.9104	nan	0.9104	0.0	0.9104

Framework versions

• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1