Vit - Cifar100 Open-source Image Classification Model, Fine-tuned on Cifar100 with a Classification Accuracy of 89.85%

Vit Cifar100

Developed by Ahmed9275

An image classification model fine-tuned on the Cifar100 dataset based on Google's ViT base model, achieving an accuracy of 89.85%

Image Classification

Transformers

Open Source License:Apache-2.0 #ViT fine-tuned model #Cifar100 high accuracy #Few-shot image classification

Downloads 920

Release Time : 5/18/2022

Model Overview

This model is an image classification model based on the Vision Transformer (ViT) architecture, specifically fine-tuned and optimized for the Cifar100 dataset, suitable for general image classification tasks

Model Features

High accuracy

Achieves 89.85% classification accuracy on the Cifar100 test set

Based on ViT architecture

Utilizes the Vision Transformer architecture, processing images with self-attention mechanisms

Efficient fine-tuning

Requires only 4 training epochs to achieve good performance

Model Capabilities

Image classification

Multi-category recognition

Use Cases

Computer vision

General image classification

Classifies images from 100 categories in the Cifar100 dataset

89.85% test accuracy

Educational research

Can serve as a benchmark model for computer vision education

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.08	1.0	3125	0.6196	0.8262
0.3816	2.0	6250	0.5322	0.8555
0.1619	3.0	9375	0.4817	0.8765
0.0443	4.0	12500	0.4420	0.8985

Property	Details
Model Type	Fine - tuned Vision Transformer (ViT) for image classification
Training Data	Cifar100
Metrics	Accuracy
Results on Evaluation Set	Loss: 0.4420, Accuracy: 0.8985

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご

Vit Cifar100

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 vit-base-beans-demo-v5

🚀 Quick Start

✨ Features

🔧 Technical Details

Training hyperparameters

Training results

Framework versions

📄 License

📋 Model Information