Vit-base-DogSick Open-source Visual Classification Model - Free Deployment to Boost Image Recognition in Specific Fields

Vit Base DogSick

Developed by jungjongho

A visual classification model fine-tuned based on Google's ViT base model, suitable for domain-specific image recognition tasks

Image Classification

Transformers

Open Source License:Apache-2.0 #Visual Classification #Fine-tuned Model #F1 Optimization

Downloads 29

Release Time : 10/22/2022

Model Overview

This model is a fine-tuned version of Google's ViT-base-patch16-224-in21k model, primarily used for image classification tasks. It demonstrates moderate accuracy and F1 scores in evaluations.

Model Features

Based on ViT Architecture

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

Medium-Scale Model

Built on the ViT-base architecture, balancing performance and computational resource requirements

Fine-Tuning Optimization

Fine-tuned on domain-specific datasets, potentially optimized for specific types of image classification tasks

Model Capabilities

Image Classification

Feature Extraction

Use Cases

Computer Vision

Domain-Specific Image Classification

Can be used for domain-specific image classification tasks such as medical image analysis, industrial quality inspection, etc.

Achieved 61% accuracy and 59.8% F1 score on the evaluation set

🚀 vit-base-DogSick

A fine - tuned version of google/vit-base-patch16-224-in21k for specific tasks.

🚀 Quick Start

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:

Loss: 1.3041
Acc: {'accuracy': 0.6102564102564103}
F1: {'f1': 0.5980148081337936}

📚 Documentation

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:

learning_rate: 4e - 05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e - 08
lr_scheduler_type: linear
num_epochs: 8
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Acc	F1
2.4055	0.61	50	2.2086	{'accuracy': 0.41794871794871796}	{'f1': 0.3246788612052483}
2.0379	1.22	100	1.9233	{'accuracy': 0.4846153846153846}	{'f1': 0.4386383497855148}
1.7287	1.83	150	1.7200	{'accuracy': 0.5256410256410257}	{'f1': 0.4806042289317683}
1.4667	2.44	200	1.6021	{'accuracy': 0.5692307692307692}	{'f1': 0.533374137436958}
1.3444	3.05	250	1.5410	{'accuracy': 0.5333333333333333}	{'f1': 0.4846134797922835}
1.1334	3.66	300	1.4674	{'accuracy': 0.5743589743589743}	{'f1': 0.5533432367508125}
1.007	4.27	350	1.4096	{'accuracy': 0.5923076923076923}	{'f1': 0.5801847507206119}
0.897	4.88	400	1.3674	{'accuracy': 0.6}	{'f1': 0.5903283954748092}
0.7326	5.49	450	1.3359	{'accuracy': 0.5923076923076923}	{'f1': 0.5793036546532927}
0.7105	6.1	500	1.3259	{'accuracy': 0.6153846153846154}	{'f1': 0.6064330281486513}
0.6164	6.71	550	1.3183	{'accuracy': 0.6102564102564103}	{'f1': 0.6014695572651212}
0.5804	7.32	600	1.3103	{'accuracy': 0.6025641025641025}	{'f1': 0.5965366941171513}
0.5313	7.93	650	1.3041	{'accuracy': 0.6102564102564103}	{'f1': 0.5980148081337936}

Framework versions

Transformers 4.23.1
Pytorch 1.12.1+cu113
Datasets 2.6.1
Tokenizers 0.13.1

📄 License

This model is licensed under the Apache 2.0 license.

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