Open-source "The Simpsons" character classification model - vit-base-patch16-224-in21k, easily identify with 95.3% accuracy

Vit Base Patch16 224 In21k Simpsons Family Members

Developed by DunnBC22

ViT-based fine-tuned model for Simpsons character classification with 95.3% accuracy

EnglishOpen Source License:Apache-2.0 #Simpsons Character Classification #High-precision ViT Model #Anime Image Recognition

Downloads 37

Release Time : 1/6/2023

Model Overview

This model is fine-tuned from Google's ViT-base model, specifically designed for recognizing and classifying different family member images from 'The Simpsons'.

Model Features

High Accuracy

Achieves 95.3% classification accuracy on the evaluation set

ViT-based Architecture

Utilizes Vision Transformer (ViT) base architecture with powerful image understanding capabilities

Multi-metric Evaluation

Provides multi-dimensional performance metrics including F1 score, recall, and precision

Model Capabilities

Image Classification

Simpsons Character Recognition

Multi-class Image Recognition

Use Cases

Entertainment Applications

Automatic Simpsons Character Classification

Automatically identifies and classifies different character images from 'The Simpsons'

95.3% accuracy

Educational Demonstrations

Computer Vision Teaching Example

Demonstrates how to fine-tune pre-trained vision models for specific classification problems

🚀 vit-base-patch16-224-in21k_Simpsons_Family_Members

This model is a fine - tuned image - classification model that can classify members of The Simpsons family. It is based on the [google/vit - base - patch16 - 224 - in21k](https://huggingface.co/google/vit - base - patch16 - 224 - in21k) model, offering high accuracy in image classification tasks.

🚀 Quick Start

This model is a fine - tuned version of [google/vit - base - patch16 - 224 - in21k](https://huggingface.co/google/vit - base - patch16 - 224 - in21k).

It achieves the following results on the evaluation set:

Loss: 0.2431
Accuracy: 0.9530
F1
- Weighted: 0.9522
- Micro: 0.9530
- Macro: 0.9521
Recall
- Weighted: 0.9530
- Micro: 0.9530
- Macro: 0.9531
Precision
- Weighted: 0.9605
- Micro: 0.9530
- Macro: 0.9601

✨ Features

Multiclass Image Classification: This is a multiclass image classification model of members of The Simpsons family.
High Accuracy: Demonstrates high accuracy in classifying Simpsons family members in the evaluation set.

📚 Documentation

Model description

This is a multiclass image classification model of members of The Simpsons family.

For more information on how it was created, check out the following link: https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/blob/main/Computer%20Vision/Image%20Classification/Multiclass%20Classification/Simpsons%20Family%20Images/Simpsons_family_with_hf_ViT.ipynb

Intended uses & limitations

This model is intended to demonstrate my ability to solve a complex problem using technology.

Training and evaluation data

Dataset Source: https://www.kaggle.com/datasets/williamu32/dataset - bart - or - homer

Sample Images From Dataset:

Sample Images

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0002
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon = 1e - 08
lr_scheduler_type: linear
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Weighted F1	Micro F1	Macro F1	Weighted Recall	Micro Recall	Macro Recall	Weighted Precision	Micro Precision	Macro Precision
1.5773	1.0	373	1.0482	0.7772	0.7263	0.7772	0.7261	0.7772	0.7772	0.7778	0.8933	0.7772	0.8922
0.1598	2.0	746	0.3902	0.9059	0.9028	0.9059	0.9026	0.9059	0.9059	0.9060	0.9224	0.9059	0.9219
0.027	3.0	1119	0.2431	0.9530	0.9522	0.9530	0.9521	0.9530	0.9530	0.9531	0.9605	0.9530	0.9601

Framework versions

Transformers 4.22.2
Pytorch 1.12.1
Datasets 2.5.2
Tokenizers 0.12.1

📄 License

This model is licensed under the Apache - 2.0 license.

Property	Details
Model Type	Fine - tuned image - classification model based on ViT
Training Data	From https://www.kaggle.com/datasets/williamu32/dataset - bart - or - homer
Metrics	Accuracy, F1, Recall, Precision

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