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: 0.1291
• Mean Iou: 0.4322
• Mean Accuracy: 0.8644
• Overall Accuracy: 0.8644
• Accuracy Unlabeled: nan
• Accuracy Tool: nan
• Accuracy Wear: 0.8644
• Iou Unlabeled: 0.0
• Iou Tool: nan
• Iou Wear: 0.8644

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
0.8371	1.82	20	0.9482	0.3285	0.9854	0.9854	nan	nan	0.9854	0.0	0.0	0.9854
0.6335	3.64	40	0.7489	0.4996	0.9992	0.9992	nan	nan	0.9992	0.0	nan	0.9992
0.5053	5.45	60	0.5400	0.4975	0.9949	0.9949	nan	nan	0.9949	0.0	nan	0.9949
0.3924	7.27	80	0.4544	0.4905	0.9810	0.9810	nan	nan	0.9810	0.0	nan	0.9810
0.3419	9.09	100	0.3840	0.4727	0.9455	0.9455	nan	nan	0.9455	0.0	nan	0.9455
0.3379	10.91	120	0.3407	0.4648	0.9296	0.9296	nan	nan	0.9296	0.0	nan	0.9296
0.2639	12.73	140	0.3495	0.4780	0.9559	0.9559	nan	nan	0.9559	0.0	nan	0.9559
0.224	14.55	160	0.2815	0.4541	0.9081	0.9081	nan	nan	0.9081	0.0	nan	0.9081
0.1725	16.36	180	0.2896	0.4599	0.9199	0.9199	nan	nan	0.9199	0.0	nan	0.9199
0.1623	18.18	200	0.2540	0.4679	0.9359	0.9359	nan	nan	0.9359	0.0	nan	0.9359
0.1724	20.0	220	0.2567	0.4702	0.9404	0.9404	nan	nan	0.9404	0.0	nan	0.9404
0.1503	21.82	240	0.1967	0.4459	0.8919	0.8919	nan	nan	0.8919	0.0	nan	0.8919
0.1189	23.64	260	0.2153	0.4617	0.9234	0.9234	nan	nan	0.9234	0.0	nan	0.9234
0.1007	25.45	280	0.1695	0.4324	0.8648	0.8648	nan	nan	0.8648	0.0	nan	0.8648
0.0921	27.27	300	0.1540	0.4346	0.8691	0.8691	nan	nan	0.8691	0.0	nan	0.8691
0.0897	29.09	320	0.1657	0.4538	0.9077	0.9077	nan	nan	0.9077	0.0	nan	0.9077
0.0814	30.91	340	0.1519	0.4374	0.8749	0.8749	nan	nan	0.8749	0.0	nan	0.8749
0.0729	32.73	360	0.1444	0.4430	0.8861	0.8861	nan	nan	0.8861	0.0	nan	0.8861
0.0892	34.55	380	0.1283	0.4106	0.8213	0.8213	nan	nan	0.8213	0.0	nan	0.8213
0.07	36.36	400	0.1442	0.4374	0.8748	0.8748	nan	nan	0.8748	0.0	nan	0.8748
0.0619	38.18	420	0.1391	0.4296	0.8592	0.8592	nan	nan	0.8592	0.0	nan	0.8592
0.0563	40.0	440	0.1283	0.4402	0.8804	0.8804	nan	nan	0.8804	0.0	nan	0.8804
0.0582	41.82	460	0.1275	0.4297	0.8595	0.8595	nan	nan	0.8595	0.0	nan	0.8595
0.0575	43.64	480	0.1341	0.4362	0.8724	0.8724	nan	nan	0.8724	0.0	nan	0.8724
0.068	45.45	500	0.1132	0.4181	0.8362	0.8362	nan	nan	0.8362	0.0	nan	0.8362
0.0595	47.27	520	0.1285	0.4316	0.8632	0.8632	nan	nan	0.8632	0.0	nan	0.8632
0.0558	49.09	540	0.1291	0.4322	0.8644	0.8644	nan	nan	0.8644	0.0	nan	0.8644

Framework versions

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