Skip to main content

The deep learning datasets provider

Project description

Data zoo

This repository provides unified access to multiple datasets.

Usage

First of all, you have to import data_provider from datazoo package:

from datazoo import data_provider

Then, you can select dataset from the list and get iterable:

# Dataset object
fashionmnist = data_provider(
	dataset='fashionmnist',	data_dir='data/fashionmnist/', split='test',
	download=True, columns=['index', 'image', 'class']
)

print('Dataset length:', len(fashionmnist))

# Iterate over samples
for i in fashionmnist:
    print(i) 

Classification

Single-label datasets

Dataset Name in data provider Number of classes Number of samples Source Auto downloading
MNIST mnist 10 60 000 / 10 000 torchvision Yes
Fashion MNIST fashionmnist 10 60 000 / 10 000 torchvision Yes
CIFAR-10 cifar10 10 50 000 / 10 000 torchvision Yes
CIFAR-100 cifar100 100 50 000 / 10 000 torchvision Yes
Indoor Scene Recognition indoor_scene_recon 67 15620 -- Yes
The Street View House Numbers (SVHN) svhn_cropped 10 73257 digits for training, 26032 digits for testing, and 531131 additional -- Yes
Linnaeus5 linnaeus5 5 classes: berry, bird, dog, flower, other (negative set) 1200 training images, 400 test images per class -- Yes
COIL-100 coil100 100 (100 objects) 7200 images -- Yes

License

This software is covered by MIT License.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
datazoo-0.0.3.tar.gz (65.4 kB) Copy SHA256 hash SHA256 Source None

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN SignalFx SignalFx Supporter DigiCert DigiCert EV certificate StatusPage StatusPage Status page