Edit model card

finetuned-affecthq

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

  • Loss: 0.8116
  • Accuracy: 0.7179
  • Precision: 0.7174
  • Recall: 0.7179
  • F1: 0.7167

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: 1e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 17
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision Recall F1
1.5413 1.0 174 1.4810 0.4898 0.4867 0.4898 0.4409
1.0367 2.0 348 1.0571 0.6155 0.6172 0.6155 0.6041
0.9534 3.0 522 0.9673 0.6475 0.6476 0.6475 0.6375
0.8532 4.0 696 0.9056 0.6748 0.6710 0.6748 0.6704
0.8211 5.0 870 0.8707 0.6903 0.6912 0.6903 0.6836
0.7797 6.0 1044 0.8472 0.7050 0.7050 0.7050 0.7019
0.7816 7.0 1218 0.8298 0.7111 0.7099 0.7111 0.7096
0.7135 8.0 1392 0.8186 0.7111 0.7116 0.7111 0.7105
0.6697 9.0 1566 0.8143 0.7140 0.7124 0.7140 0.7126
0.6765 10.0 1740 0.8116 0.7179 0.7174 0.7179 0.7167

Framework versions

  • Transformers 4.27.0.dev0
  • Pytorch 1.13.1+cu116
  • Datasets 2.9.0
  • Tokenizers 0.13.2
Downloads last month
2
Inference API
Drag image file here or click to browse from your device
This model can be loaded on Inference API (serverless).

Evaluation results