vit-base-patch16-224-in21k-lung_and_colon

<!-- This model card has been generated automatically according to the information Keras had access to. You should probably proofread and complete it, then remove this comment. -->

Guldeniz/vit-base-patch16-224-in21k-lung_and_colon

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on Lung and Colon Histopathological Images dataset. This dataset can be reach via Kaggle. It achieves the following results on the evaluation set:

• Train Loss: 0.0088
• Train Accuracy: 1.0
• Train Top-3-accuracy: 1.0
• Validation Loss: 0.0084
• Validation Accuracy: 0.9997
• Validation Top-3-accuracy: 1.0
• Epoch: 3

Model description

The vision transformer model, trained by Google, has been fine-tuned using a lung and colon cancer image dataset consisting of a total of 25,000 images across 5 labels. The obtained results are highly promising, and the model demonstrates the ability to predict the following listed labels.

• colon_aca

• colon_n

• lung_aca

• lung_n

• lung_scc

Training hyperparameters

The following hyperparameters were used during training:

• optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 3e-05, 'decay_steps': 3325, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: float32

Training results

Framework versions

• Transformers 4.26.1
• TensorFlow 2.12.0
• Datasets 2.10.1
• Tokenizers 0.13.3