vit-fire-detection / README.md
Madhukar7559's picture
Model save
7f6a6a2
metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
model-index:
  - name: vit-fire-detection
    results: []

vit-fire-detection

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0103
  • Precision: 0.9987
  • Recall: 0.9987

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: 0.0002
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 100
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Precision Recall
0.0797 1.0 190 0.0811 0.9789 0.9775
0.0536 2.0 380 0.0205 0.9947 0.9947
0.0374 3.0 570 0.0283 0.9922 0.9921
0.0209 4.0 760 0.0046 1.0 1.0
0.0104 5.0 950 0.0128 0.9960 0.9960
0.0159 6.0 1140 0.0152 0.9947 0.9947
0.0119 7.0 1330 0.0084 0.9974 0.9974
0.0044 8.0 1520 0.0111 0.9987 0.9987
0.0077 9.0 1710 0.0094 0.9987 0.9987
0.0106 10.0 1900 0.0103 0.9987 0.9987

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu121
  • Tokenizers 0.15.0