Edit model card

segformer-b0-finetuned-segments-SixrayKnife8-21-2024_saad2

This model is a fine-tuned version of nvidia/mit-b0 on the saad7489/SixraygunTest dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3358
  • Mean Iou: 0.7746
  • Mean Accuracy: 0.8963
  • Overall Accuracy: 0.9817
  • Accuracy Bkg: 0.9889
  • Accuracy Knife: 0.8285
  • Accuracy Gun: 0.8716
  • Iou Bkg: 0.9833
  • Iou Knife: 0.6954
  • Iou Gun: 0.6451

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: 20
  • eval_batch_size: 20
  • 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 Bkg Accuracy Knife Accuracy Gun Iou Bkg Iou Knife Iou Gun
0.7888 6.6667 20 0.8695 0.5702 0.8463 0.9402 0.9482 0.7648 0.8257 0.9426 0.4598 0.3081
0.5585 13.3333 40 0.5529 0.6147 0.7305 0.9648 0.9847 0.5317 0.6751 0.9665 0.4745 0.4032
0.4632 20.0 60 0.4458 0.6892 0.7884 0.9740 0.9891 0.6932 0.6829 0.9754 0.5996 0.4926
0.4117 26.6667 80 0.4212 0.7226 0.8764 0.9748 0.9831 0.8038 0.8423 0.9761 0.6472 0.5445
0.3782 33.3333 100 0.3883 0.7500 0.8957 0.9785 0.9856 0.8198 0.8818 0.9802 0.6733 0.5965
0.3489 40.0 120 0.3564 0.7642 0.9014 0.9802 0.9869 0.8332 0.8839 0.9819 0.6855 0.6251
0.3454 46.6667 140 0.3358 0.7746 0.8963 0.9817 0.9889 0.8285 0.8716 0.9833 0.6954 0.6451

Framework versions

  • Transformers 4.42.4
  • Pytorch 2.3.1+cu121
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
8
Safetensors
Model size
3.72M params
Tensor type
F32
·
Inference Examples
Unable to determine this model's library. Check the docs .

Model tree for saad7489/segformer-b0-finetuned-segments-SixrayKnife8-21-2024_saad2

Base model

nvidia/mit-b0
Finetuned
(319)
this model