Hugging Face's logo

varcoder
/
segcrack9k_conglomerate_segformer

Transformers
PyTorch
TensorBoard
segformer
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
segcrack9k_conglomerate_segformer / README.md
varcoder's picture
varcoder
update model card README.md
162639f
preview code
|
raw
history blame
9.72 kB
metadata
license: other
base_model: nvidia/mit-b5
tags:
  - generated_from_trainer
model-index:
  - name: segcrack9k_conglomerate_segformer
    results: []

segcrack9k_conglomerate_segformer

This model is a fine-tuned version of nvidia/mit-b5 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0857
  • Mean Iou: 0.0010
  • Mean Accuracy: 0.0021
  • Overall Accuracy: 0.0021
  • Accuracy Background: nan
  • Accuracy Crack: 0.0021
  • Iou Background: 0.0
  • Iou Crack: 0.0021

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: 8e-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: 1

Training results

Training Loss Epoch Step Validation Loss Mean Iou Mean Accuracy Overall Accuracy Accuracy Background Accuracy Crack Iou Background Iou Crack
0.0716 0.02 100 0.1132 0.0 0.0 0.0 nan 0.0 0.0 0.0
0.0708 0.04 200 0.1006 0.0001 0.0003 0.0003 nan 0.0003 0.0 0.0003
0.1661 0.06 300 0.1042 0.0 0.0 0.0 nan 0.0 0.0 0.0
0.0601 0.08 400 0.1005 0.0 0.0 0.0 nan 0.0 0.0 0.0
0.1034 0.1 500 0.0980 0.0237 0.0474 0.0474 nan 0.0474 0.0 0.0474
0.0581 0.12 600 0.0965 0.0003 0.0005 0.0005 nan 0.0005 0.0 0.0005
0.0561 0.14 700 0.1023 0.0038 0.0075 0.0075 nan 0.0075 0.0 0.0075
0.1034 0.16 800 0.0956 0.0002 0.0003 0.0003 nan 0.0003 0.0 0.0003
0.1341 0.18 900 0.0985 0.0185 0.0369 0.0369 nan 0.0369 0.0 0.0369
0.1988 0.2 1000 0.0946 0.0059 0.0118 0.0118 nan 0.0118 0.0 0.0118
0.0378 0.22 1100 0.0945 0.1402 0.2804 0.2804 nan 0.2804 0.0 0.2804
0.0582 0.24 1200 0.0907 0.0488 0.0976 0.0976 nan 0.0976 0.0 0.0976
0.1464 0.26 1300 0.0971 0.1701 0.3401 0.3401 nan 0.3401 0.0 0.3401
0.0601 0.28 1400 0.0893 0.0222 0.0444 0.0444 nan 0.0444 0.0 0.0444
0.0855 0.3 1500 0.0910 0.0307 0.0613 0.0613 nan 0.0613 0.0 0.0613
0.1167 0.32 1600 0.0895 0.0143 0.0286 0.0286 nan 0.0286 0.0 0.0286
0.0641 0.34 1700 0.0918 0.0073 0.0145 0.0145 nan 0.0145 0.0 0.0145
0.0621 0.36 1800 0.0927 0.0181 0.0363 0.0363 nan 0.0363 0.0 0.0363
0.0364 0.38 1900 0.0884 0.1397 0.2794 0.2794 nan 0.2794 0.0 0.2794
0.1394 0.4 2000 0.0903 0.0000 0.0000 0.0000 nan 0.0000 0.0 0.0000
0.0187 0.42 2100 0.0914 0.0124 0.0248 0.0248 nan 0.0248 0.0 0.0248
0.1842 0.44 2200 0.0908 0.0045 0.0090 0.0090 nan 0.0090 0.0 0.0090
0.0847 0.46 2300 0.0896 0.0031 0.0062 0.0062 nan 0.0062 0.0 0.0062
0.0556 0.48 2400 0.0871 0.0016 0.0033 0.0033 nan 0.0033 0.0 0.0033
0.0454 0.51 2500 0.0896 0.0005 0.0010 0.0010 nan 0.0010 0.0 0.0010
0.1411 0.53 2600 0.0876 0.0095 0.0190 0.0190 nan 0.0190 0.0 0.0190
0.1044 0.55 2700 0.0936 0.0001 0.0002 0.0002 nan 0.0002 0.0 0.0002
0.1299 0.57 2800 0.0938 0.0008 0.0017 0.0017 nan 0.0017 0.0 0.0017
0.0909 0.59 2900 0.0877 0.0012 0.0024 0.0024 nan 0.0024 0.0 0.0024
0.0981 0.61 3000 0.0914 0.0012 0.0024 0.0024 nan 0.0024 0.0 0.0024
0.0905 0.63 3100 0.0880 0.0077 0.0153 0.0153 nan 0.0153 0.0 0.0153
0.2111 0.65 3200 0.0877 0.0000 0.0001 0.0001 nan 0.0001 0.0 0.0001
0.3218 0.67 3300 0.0860 0.0036 0.0072 0.0072 nan 0.0072 0.0 0.0072
0.1134 0.69 3400 0.0864 0.0075 0.0151 0.0151 nan 0.0151 0.0 0.0151
0.2184 0.71 3500 0.0907 0.0000 0.0000 0.0000 nan 0.0000 0.0 0.0000
0.1779 0.73 3600 0.0877 0.0029 0.0059 0.0059 nan 0.0059 0.0 0.0059
0.3664 0.75 3700 0.0878 0.0001 0.0001 0.0001 nan 0.0001 0.0 0.0001
0.0365 0.77 3800 0.0870 0.0000 0.0000 0.0000 nan 0.0000 0.0 0.0000
0.0591 0.79 3900 0.0877 0.0000 0.0001 0.0001 nan 0.0001 0.0 0.0001
0.0719 0.81 4000 0.0871 0.0004 0.0008 0.0008 nan 0.0008 0.0 0.0008
0.0402 0.83 4100 0.0874 0.0011 0.0022 0.0022 nan 0.0022 0.0 0.0022
0.0814 0.85 4200 0.0887 0.0008 0.0017 0.0017 nan 0.0017 0.0 0.0017
0.0485 0.87 4300 0.0871 0.0025 0.0050 0.0050 nan 0.0050 0.0 0.0050
0.0487 0.89 4400 0.0864 0.0004 0.0007 0.0007 nan 0.0007 0.0 0.0007
0.0689 0.91 4500 0.0859 0.0002 0.0004 0.0004 nan 0.0004 0.0 0.0004
0.0782 0.93 4600 0.0858 0.0018 0.0036 0.0036 nan 0.0036 0.0 0.0036
0.2153 0.95 4700 0.0855 0.0004 0.0008 0.0008 nan 0.0008 0.0 0.0008
0.1974 0.97 4800 0.0860 0.0004 0.0009 0.0009 nan 0.0009 0.0 0.0009
0.0184 0.99 4900 0.0857 0.0010 0.0021 0.0021 nan 0.0021 0.0 0.0021

Framework versions

  • Transformers 4.31.0
  • Pytorch 2.0.1+cu118
  • Datasets 2.13.1
  • Tokenizers 0.13.3