ResNet26 Open-source Image Classification Model - Free Deployment for Efficient Image Category Recognition

Resnet26

Developed by glasses

ResNet26 is an image classification model based on the deep residual learning architecture, a variant in the ResNet series.

Image Classification

Transformers

Open Source License:Apache-2.0 #Deep Residual Network #Image Classification Optimization #Modular Design

Downloads 14

Release Time : 3/2/2022

Model Overview

ResNet26 implements the ResNet architecture proposed in the paper 'Deep Residual Learning for Image Recognition,' primarily used for image classification tasks. The model addresses the vanishing gradient problem in deep neural networks through residual connections and supports multiple variants and custom configurations.

Model Features

Multiple Variant Support

Provides multiple depth variants from ResNet18 to ResNet200, as well as improved 'd' series variants.

Highly Customizable

Supports custom stem structures, block modules, and shortcut connection methods to flexibly adapt to different needs.

Feature Extraction Capability

Easily extracts features from various layers, suitable for transfer learning and feature engineering.

Model Capabilities

Image Classification

Feature Extraction

Use Cases

Computer Vision

ImageNet Classification

Performs 1000-class image classification on the ImageNet dataset

Transfer Learning

Used as a pre-trained model for transfer learning in other visual tasks

🚀 ResNet26

This project implements the ResNet model proposed in Deep Residual Learning for Image Recognition, which is used for image classification tasks.

🚀 Quick Start

The ResNet model implemented in this project provides multiple variants, allowing users to choose according to their needs.

✨ Features

Implement various ResNet variants, including resnet18, resnet26, resnet34, resnet50, resnet101, resnet152, and resnet200.
Support variants (d) proposed in Bag of Tricks for Image Classification with Convolutional Neural Networks <https://arxiv.org/pdf/1812.01187.pdf>_.
Allows users to customize models by changing stem and block.

💻 Usage Examples

Basic Usage

ResNet.resnet18()
ResNet.resnet26()
ResNet.resnet34()
ResNet.resnet50()
ResNet.resnet101()
ResNet.resnet152()
ResNet.resnet200()

# Variants (d) proposed in `Bag of Tricks for Image Classification with Convolutional Neural Networks <https://arxiv.org/pdf/1812.01187.pdf`_
ResNet.resnet26d()
ResNet.resnet34d()
ResNet.resnet50d()
# You can construct your own one by chaning `stem` and `block`
resnet101d = ResNet.resnet101(stem=ResNetStemC, block=partial(ResNetBottleneckBlock, shortcut=ResNetShorcutD))

Advanced Usage

# change activation
ResNet.resnet18(activation = nn.SELU)
# change number of classes (default is 1000 )
ResNet.resnet18(n_classes=100)
# pass a different block
ResNet.resnet18(block=SENetBasicBlock)
# change the steam
model = ResNet.resnet18(stem=ResNetStemC)
# change shortcut
model = ResNet.resnet18(block=partial(ResNetBasicBlock, shortcut=ResNetShorcutD))
# store each feature
x = torch.rand((1, 3, 224, 224))
# get features
model = ResNet.resnet18()
# first call .features, this will activate the forward hooks and tells the model you'll like to get the features
model.encoder.features
model(torch.randn((1,3,224,224)))
# get the features from the encoder
features = model.encoder.features
print([x.shape for x in features])
#[torch.Size([1, 64, 112, 112]), torch.Size([1, 64, 56, 56]), torch.Size([1, 128, 28, 28]), torch.Size([1, 256, 14, 14])]

📄 License

This project is licensed under the Apache-2.0 license.

Property	Details
Model Type	Image Classification
Training Data	ImageNet

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