ryan_model3272024 / README.md
rshrott's picture
🍻 cheers
2a7c83b verified
|
raw
history blame
2.68 kB
metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - image-classification
  - generated_from_trainer
model-index:
  - name: ryan_model3272024
    results: []

ryan_model3272024

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

  • Loss: 0.2552
  • Na Accuracy: 0.95
  • Ordinal Accuracy: 0.6267
  • Ordinal Mae: 1.1586

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: 16
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 4
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Na Accuracy Ordinal Accuracy Ordinal Mae
0.3853 0.32 100 0.3272 0.924 0.52 1.2106
0.3396 0.64 200 0.2741 0.94 0.5644 1.1640
0.2075 0.96 300 0.2772 0.946 0.5933 1.1942
0.196 1.28 400 0.2738 0.95 0.6133 1.1984
0.2228 1.6 500 0.2685 0.956 0.62 1.1989
0.1816 1.92 600 0.2552 0.95 0.6267 1.1586
0.0682 2.24 700 0.2721 0.952 0.6578 1.1558
0.0795 2.56 800 0.2754 0.948 0.6333 1.1599
0.1367 2.88 900 0.2953 0.946 0.64 1.1667
0.0387 3.19 1000 0.2923 0.944 0.6378 1.2025
0.0293 3.51 1100 0.2885 0.948 0.6644 1.1666
0.0286 3.83 1200 0.2868 0.95 0.6711 1.1626

Framework versions

  • Transformers 4.39.1
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2