license: apache-2.0
tags:

• 图像分类
• pytorch
• onnx
  datasets:
• pyronear/openfire

ReXNet-1.3x 模型

该模型基于野火二分类数据集预训练（即将公开）。ReXNet架构源自这篇论文。

模型描述

作者的核心创新是在残差块中加入定制化的Squeeze-Excitation层，以消除通道冗余问题。

安装指南

环境要求

需安装Python 3.6（或更高版本）及pip或conda。

稳定版安装

可通过pypi安装最新稳定版：

pip install pyrovision  

或通过conda安装：

conda install -c pyronear pyrovision  

开发者模式

如需体验未发布的最新功能，可从源码安装（需先安装Git）：

git clone https://github.com/pyronear/pyro-vision.git  
pip install -e pyro-vision/.  

使用说明

from PIL import Image  
from torchvision.transforms import Compose, ConvertImageDtype, Normalize, PILToTensor, Resize  
from torchvision.transforms.functional import InterpolationMode  
from pyrovision.models import model_from_hf_hub  

model = model_from_hf_hub("pyronear/rexnet1_3x").eval()  

img = Image.open(图片路径).convert("RGB")  

# 预处理  
config = model.default_cfg  
transform = Compose([  
    Resize(config['input_shape'][1:], interpolation=InterpolationMode.BILINEAR),  
    PILToTensor(),  
    ConvertImageDtype(torch.float32),  
    Normalize(config['mean'], config['std'])  
])  

input_tensor = transform(img).unsqueeze(0)  

# 推理  
with torch.inference_mode():  
    output = model(input_tensor)  
probs = output.squeeze(0).softmax(dim=0)  

引用文献

原始论文：

@article{DBLP:journals/corr/abs-2007-00992,  
  author    = {Dongyoon Han and  
               Sangdoo Yun and  
               Byeongho Heo and  
               Young Joon Yoo},  
  title     = {ReXNet: Diminishing Representational Bottleneck on Convolutional Neural  
               Network},  
  journal   = {CoRR},  
  volume    = {abs/2007.00992},  
  year      = {2020},  
  url       = {https://arxiv.org/abs/2007.00992},  
  eprinttype = {arXiv},  
  eprint    = {2007.00992},  
  timestamp = {Mon, 06 Jul 2020 15:26:01 +0200},  
  biburl    = {https://dblp.org/rec/journals/corr/abs-2007-00992.bib},  
  bibsource = {dblp computer science bibliography, https://dblp.org}  
}  

实现源码：

@software{Fernandez_Holocron_2020,  
author = {Fernandez, François-Guillaume},  
month = {5},  
title = {{Holocron}},  
url = {https://github.com/frgfm/Holocron},  
year = {2020}  
}