Hugging Face's logo

HorcruxNo13
/
segformer-b0-finetuned-segments-toolwear

Transformers
PyTorch
TensorBoard
Safetensors
segformer
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
segformer-b0-finetuned-segments-toolwear / README.md
HorcruxNo13's picture
HorcruxNo13
update model card README.md
95f02f1
preview code
|
raw
history blame
6.84 kB
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_edges dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7517
  • Mean Iou: 0.3530
  • Mean Accuracy: 0.7066
  • Overall Accuracy: 0.7444
  • Accuracy Unlabeled: nan
  • Accuracy Tool: 0.6653
  • Accuracy Wear: 0.7480
  • Iou Unlabeled: 0.0
  • Iou Tool: 0.3188
  • Iou Wear: 0.7403

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
0.8264 1.82 20 0.9929 0.3016 0.5119 0.6940 nan 0.3130 0.7109 0.0 0.2149 0.6899
0.566 3.64 40 0.8390 0.3172 0.6658 0.6561 nan 0.6765 0.6552 0.0 0.3052 0.6466
0.5515 5.45 60 0.7996 0.3015 0.7085 0.5831 nan 0.8455 0.5715 0.0 0.3365 0.5680
0.496 7.27 80 0.7495 0.3370 0.7783 0.6771 nan 0.8889 0.6676 0.0 0.3465 0.6645
0.4945 9.09 100 0.7214 0.3106 0.6966 0.6150 nan 0.7858 0.6074 0.0 0.3294 0.6025
0.4392 10.91 120 0.7105 0.3012 0.7519 0.5990 nan 0.9191 0.5848 0.0 0.3198 0.5839
0.3211 12.73 140 0.7570 0.3470 0.7008 0.7352 nan 0.6632 0.7384 0.0 0.3116 0.7292
0.2289 14.55 160 0.9477 0.3748 0.7214 0.7566 nan 0.6830 0.7598 0.0 0.3718 0.7527
0.4674 16.36 180 0.8172 0.3637 0.7442 0.7533 nan 0.7344 0.7541 0.0 0.3437 0.7476
0.3226 18.18 200 0.8199 0.3238 0.7286 0.6845 nan 0.7769 0.6804 0.0 0.2939 0.6777
0.1706 20.0 220 0.7336 0.3410 0.6894 0.7096 nan 0.6673 0.7115 0.0 0.3185 0.7044
0.2786 21.82 240 0.9254 0.3662 0.7577 0.7864 nan 0.7264 0.7891 0.0 0.3164 0.7821
0.1685 23.64 260 0.8291 0.3435 0.7685 0.7294 nan 0.8113 0.7258 0.0 0.3082 0.7224
0.1649 25.45 280 0.7200 0.3303 0.7133 0.6593 nan 0.7723 0.6543 0.0 0.3394 0.6516
0.1481 27.27 300 0.8155 0.3531 0.7558 0.7434 nan 0.7695 0.7422 0.0 0.3206 0.7385
0.1476 29.09 320 0.7374 0.3455 0.6734 0.7252 nan 0.6169 0.7300 0.0 0.3153 0.7211
0.2284 30.91 340 0.7254 0.3265 0.6989 0.6766 nan 0.7233 0.6745 0.0 0.3099 0.6695
0.1212 32.73 360 0.8022 0.3591 0.7252 0.7662 nan 0.6804 0.7700 0.0 0.3153 0.7620
0.1284 34.55 380 0.7345 0.3449 0.7044 0.7331 nan 0.6731 0.7357 0.0 0.3062 0.7284
0.1685 36.36 400 0.7581 0.3275 0.7357 0.6991 nan 0.7757 0.6957 0.0 0.2910 0.6915
0.1018 38.18 420 0.7303 0.3401 0.6575 0.7173 nan 0.5921 0.7228 0.0 0.3069 0.7133
0.1405 40.0 440 0.7375 0.3555 0.7301 0.7475 nan 0.7111 0.7491 0.0 0.3234 0.7431
0.08 41.82 460 0.7449 0.3561 0.7047 0.7457 nan 0.6598 0.7495 0.0 0.3265 0.7417
0.1311 43.64 480 0.7680 0.3552 0.7205 0.7444 nan 0.6945 0.7466 0.0 0.3257 0.7398
0.1235 45.45 500 0.7589 0.3523 0.7117 0.7398 nan 0.6811 0.7424 0.0 0.3218 0.7352
0.1169 47.27 520 0.7676 0.3535 0.6952 0.7529 nan 0.6320 0.7583 0.0 0.3110 0.7494
0.14 49.09 540 0.7517 0.3530 0.7066 0.7444 nan 0.6653 0.7480 0.0 0.3188 0.7403

Framework versions

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