Edit model card

ansilmbabl/vit-base-patch16-224-in21k-Cards

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:

  • Train Loss: 0.0861
  • Train Accuracy: 0.9775
  • Train Top-3-accuracy: 0.9991
  • Validation Loss: 0.6749
  • Validation Accuracy: 0.8113
  • Validation Top-3-accuracy: 0.9740
  • Epoch: 5

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:

  • optimizer: {'inner_optimizer': {'module': 'transformers.optimization_tf', 'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 3e-05, 'decay_steps': 53200, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'decay': 0.0, 'beta_1': 0.8999999761581421, 'beta_2': 0.9990000128746033, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}, 'registered_name': 'AdamWeightDecay'}, 'dynamic': True, 'initial_scale': 32768.0, 'dynamic_growth_steps': 2000}
  • training_precision: mixed_float16

Training results

Train Loss Train Accuracy Train Top-3-accuracy Validation Loss Validation Accuracy Validation Top-3-accuracy Epoch
1.3188 0.6043 0.8822 0.8883 0.7130 0.9537 0
0.6864 0.7853 0.9705 0.6807 0.7647 0.9650 1
0.3966 0.8815 0.9896 0.5364 0.8120 0.9777 2
0.2310 0.9338 0.9964 0.4965 0.8427 0.9817 3
0.1333 0.9649 0.9981 0.4799 0.8513 0.9837 4
0.0861 0.9775 0.9991 0.6749 0.8113 0.9740 5

Framework versions

  • Transformers 4.41.2
  • TensorFlow 2.14.0
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
4
Inference API
Drag image file here or click to browse from your device
This model can be loaded on Inference API (serverless).

Finetuned from