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: 0.2168
• Mean Iou: 0.3007
• Mean Accuracy: 0.6014
• Overall Accuracy: 0.6014
• Accuracy Unlabeled: nan
• Accuracy Tool: nan
• Accuracy Wear: 0.6014
• Iou Unlabeled: 0.0
• Iou Tool: nan
• Iou Wear: 0.6014

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: 25

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
0.9147	1.82	20	0.9308	0.3010	0.9029	0.9029	nan	nan	0.9029	0.0	0.0	0.9029
0.6788	3.64	40	0.6521	0.2589	0.7768	0.7768	nan	nan	0.7768	0.0	0.0	0.7768
0.4559	5.45	60	0.4600	0.2663	0.7989	0.7989	nan	nan	0.7989	0.0	0.0	0.7989
0.3799	7.27	80	0.3767	0.2061	0.6182	0.6182	nan	nan	0.6182	0.0	0.0	0.6182
0.4438	9.09	100	0.3259	0.3479	0.6958	0.6958	nan	nan	0.6958	0.0	nan	0.6958
0.3534	10.91	120	0.3008	0.3057	0.6114	0.6114	nan	nan	0.6114	0.0	nan	0.6114
0.3332	12.73	140	0.2805	0.3631	0.7261	0.7261	nan	nan	0.7261	0.0	nan	0.7261
0.2543	14.55	160	0.2659	0.2927	0.5853	0.5853	nan	nan	0.5853	0.0	nan	0.5853
0.2746	16.36	180	0.2324	0.2724	0.5449	0.5449	nan	nan	0.5449	0.0	nan	0.5449
0.2532	18.18	200	0.2409	0.3597	0.7194	0.7194	nan	nan	0.7194	0.0	nan	0.7194
0.2353	20.0	220	0.2369	0.3070	0.6139	0.6139	nan	nan	0.6139	0.0	nan	0.6139
0.2192	21.82	240	0.2210	0.3041	0.6083	0.6083	nan	nan	0.6083	0.0	nan	0.6083
0.2469	23.64	260	0.2168	0.3007	0.6014	0.6014	nan	nan	0.6014	0.0	nan	0.6014

Framework versions

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