Vit Base Patch16 224 In21k Eurosat
This model is a fine-tuned Vision Transformer based on google/vit-base-patch16-224-in21k on an unknown dataset, primarily used for image classification tasks.
Downloads 23
Release Time : 4/15/2022
Model Overview
This is an image classification model based on the Vision Transformer architecture, fine-tuned to perform exceptionally well on the EuroSat dataset.
Model Features
High accuracy
Achieves 98.39% accuracy on the validation set and 100% Top-3 accuracy
Pre-trained model based
Fine-tuned from the google/vit-base-patch16-224-in21k pre-trained model
Efficient training
Requires only 3 epochs of training to achieve excellent performance
Model Capabilities
Image classification
Multi-class recognition
Use Cases
Remote sensing image analysis
Satellite image classification
Classify satellite images in the EuroSat dataset
Achieves 98.39% classification accuracy
Geographic information systems
Land use classification
Identify different land cover types
Featured Recommended AI Models
Š 2025AIbase