Dataset card for SVHN

The Street View House Numbers (SVHN) dataset is a real-world image dataset developed and designed for machine learning and object recognition algorithms, and is characterized by low data preprocessing and formatting requirements. Similar to MNIST, SVHN contains images of small cropped numbers, but in terms of labeled data, SVHN is an order of magnitude larger than MNIST, comprising over 600,000 digital images. Unlike MNIST, SVHN deals with a much more challenging and unsolved real-world problem - recognizing numbers and strings of numbers in natural scene images. The dataset is derived from house numbers captured in Google Street View images.

Maintenance

git clone git@hf.co:datasets/MuGeminorum/svhn
cd svhn

Usage

import os
import zipfile
import requests

def download_file(url, save_path):
    response = requests.get(url, stream=True)
    with open(save_path, 'wb') as file:
        for chunk in response.iter_content(chunk_size=1024):
            if chunk:
                file.write(chunk)

def unzip(zip_file_path, extract_to):
    with zipfile.ZipFile(zip_file_path, 'r') as zip_ref:
        for member in zip_ref.infolist():
            zip_ref.extract(member, extract_to)

if not os.path.exists('./data.zip'):
    download_file(
        'https://www.modelscope.cn/datasets/MuGeminorum/svhn/resolve/master/data.zip',
        'data.zip'
    )

if not os.path.exists('./data'):
    unzip('data.zip', './')

Mirror

https://www.modelscope.cn/datasets/MuGeminorum/svhn

Reference

[1] The Street View House Numbers (SVHN) Dataset
[2] https://github.com/MuGeminorum/SVHN-Recognition