README.md

metadata

license: other
tags:
  - generated_from_trainer
model-index:
  - name: segformer-b0-finetuned-segments-toolwear
    results: []

segformer-b0-finetuned-segments-toolwear

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

• Loss: 0.0517
• Mean Iou: 0.3741
• Mean Accuracy: 0.7482
• Overall Accuracy: 0.7482
• Accuracy Unlabeled: nan
• Accuracy Tool: nan
• Accuracy Wear: 0.7482
• Iou Unlabeled: 0.0
• Iou Tool: nan
• Iou Wear: 0.7482

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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Tool	Accuracy Wear	Iou Unlabeled	Iou Tool	Iou Wear
0.8497	1.82	20	0.8647	0.4917	0.9834	0.9834	nan	nan	0.9834	0.0	nan	0.9834
0.6095	3.64	40	0.5158	0.4642	0.9283	0.9283	nan	nan	0.9283	0.0	nan	0.9283
0.4377	5.45	60	0.4200	0.4646	0.9291	0.9291	nan	nan	0.9291	0.0	nan	0.9291
0.3756	7.27	80	0.3535	0.4780	0.9560	0.9560	nan	nan	0.9560	0.0	nan	0.9560
0.4256	9.09	100	0.2951	0.4873	0.9746	0.9746	nan	nan	0.9746	0.0	nan	0.9746
0.2748	10.91	120	0.2500	0.4817	0.9634	0.9634	nan	nan	0.9634	0.0	nan	0.9634
0.2347	12.73	140	0.2000	0.4065	0.8129	0.8129	nan	nan	0.8129	0.0	nan	0.8129
0.1777	14.55	160	0.1651	0.4340	0.8680	0.8680	nan	nan	0.8680	0.0	nan	0.8680
0.186	16.36	180	0.1530	0.4211	0.8422	0.8422	nan	nan	0.8422	0.0	nan	0.8422
0.1652	18.18	200	0.1143	0.4304	0.8608	0.8608	nan	nan	0.8608	0.0	nan	0.8608
0.1227	20.0	220	0.1436	0.4838	0.9676	0.9676	nan	nan	0.9676	0.0	nan	0.9676
0.1111	21.82	240	0.1014	0.3994	0.7988	0.7988	nan	nan	0.7988	0.0	nan	0.7988
0.0989	23.64	260	0.0914	0.3574	0.7147	0.7147	nan	nan	0.7147	0.0	nan	0.7147
0.1051	25.45	280	0.0871	0.2844	0.5689	0.5689	nan	nan	0.5689	0.0	nan	0.5689
0.0975	27.27	300	0.0679	0.3893	0.7786	0.7786	nan	nan	0.7786	0.0	nan	0.7786
0.0928	29.09	320	0.0723	0.4241	0.8483	0.8483	nan	nan	0.8483	0.0	nan	0.8483
0.0673	30.91	340	0.0653	0.3628	0.7255	0.7255	nan	nan	0.7255	0.0	nan	0.7255
0.0652	32.73	360	0.0641	0.4023	0.8047	0.8047	nan	nan	0.8047	0.0	nan	0.8047
0.0912	34.55	380	0.0734	0.4453	0.8906	0.8906	nan	nan	0.8906	0.0	nan	0.8906
0.0682	36.36	400	0.0609	0.3322	0.6644	0.6644	nan	nan	0.6644	0.0	nan	0.6644
0.0737	38.18	420	0.0619	0.4053	0.8107	0.8107	nan	nan	0.8107	0.0	nan	0.8107
0.06	40.0	440	0.0564	0.3593	0.7186	0.7186	nan	nan	0.7186	0.0	nan	0.7186
0.0555	41.82	460	0.0562	0.4025	0.8050	0.8050	nan	nan	0.8050	0.0	nan	0.8050
0.063	43.64	480	0.0550	0.3945	0.7891	0.7891	nan	nan	0.7891	0.0	nan	0.7891
0.0641	45.45	500	0.0554	0.4032	0.8065	0.8065	nan	nan	0.8065	0.0	nan	0.8065
0.0739	47.27	520	0.0549	0.3880	0.7760	0.7760	nan	nan	0.7760	0.0	nan	0.7760
0.0684	49.09	540	0.0517	0.3741	0.7482	0.7482	nan	nan	0.7482	0.0	nan	0.7482

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3