README.md

metadata

license: other
tags:
  - vision
  - image-segmentation
  - 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 the HorcruxNo13/toolwear_segmentsai dataset. It achieves the following results on the evaluation set:

• Loss: 0.2940
• Mean Iou: 0.4104
• Mean Accuracy: 0.8207
• Overall Accuracy: 0.8207
• Accuracy Unlabeled: nan
• Accuracy Tool: nan
• Accuracy Wear: 0.8207
• Iou Unlabeled: 0.0
• Iou Tool: nan
• Iou Wear: 0.8207

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
0.843	1.82	20	0.8832	0.4637	0.9274	0.9274	nan	nan	0.9274	0.0	nan	0.9274
0.6849	3.64	40	0.5914	0.4361	0.8722	0.8722	nan	nan	0.8722	0.0	nan	0.8722
0.5085	5.45	60	0.5178	0.4628	0.9256	0.9256	nan	nan	0.9256	0.0	nan	0.9256
0.4027	7.27	80	0.5099	0.4598	0.9195	0.9195	nan	nan	0.9195	0.0	nan	0.9195
0.3536	9.09	100	0.4262	0.4365	0.8730	0.8730	nan	nan	0.8730	0.0	nan	0.8730
0.3657	10.91	120	0.3891	0.4228	0.8457	0.8457	nan	nan	0.8457	0.0	nan	0.8457
0.3234	12.73	140	0.4221	0.4377	0.8754	0.8754	nan	nan	0.8754	0.0	nan	0.8754
0.2874	14.55	160	0.3355	0.4098	0.8197	0.8197	nan	nan	0.8197	0.0	nan	0.8197
0.2335	16.36	180	0.3570	0.4266	0.8531	0.8531	nan	nan	0.8531	0.0	nan	0.8531
0.2167	18.18	200	0.3238	0.4404	0.8808	0.8808	nan	nan	0.8808	0.0	nan	0.8808
0.2201	20.0	220	0.3103	0.4185	0.8370	0.8370	nan	nan	0.8370	0.0	nan	0.8370
0.205	21.82	240	0.2881	0.4115	0.8230	0.8230	nan	nan	0.8230	0.0	nan	0.8230
0.241	23.64	260	0.2940	0.4104	0.8207	0.8207	nan	nan	0.8207	0.0	nan	0.8207

Framework versions

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