DALL-E-Cats is a dataset meant to produce a synthetic animal dataset. This is a successor to DALL-E-Dogs. DALL-E-Dogs and DALL-E-Cats will be fed into an image classifier to see how it performs. This is under the BirdL-AirL License.

This statistic shows the results of a survey conducted in the United States in 2017 on pets. Some 51 percent of the respondents stated that they prefer dogs.The Survey Data Table for the Statista survey pets in the U.S. 2017 contains the complete tables for the survey including various column headings.

A large set of images of cats and dogs.

Homepage: https://www.microsoft.com/en-us/download/details.aspx?id=54765

Source code: tfds.image_classification.CatsVsDogs

Versions:

4.0.0 (default): New split API (https://tensorflow.org/datasets/splits) Download size: 786.68 MiB

Source: https://www.tensorflow.org/datasets/catalog/cats_vs_dogs

Dataset Card for Cats Vs. Dogs

  Dataset Summary

A large set of images of cats and dogs. There are 1738 corrupted images that are dropped. This dataset is part of a now-closed Kaggle competition and represents a subset of the so-called Asirra dataset. From the competition page:

The Asirra data set Web services are often protected with a challenge that's supposed to be easy for people to solve, but difficult for computers. Such a challenge is often called a CAPTCHA… See the full description on the dataset page: https://huggingface.co/datasets/microsoft/cats_vs_dogs.

The US Cat Food Market is segmented by Pet Food Product (Food, Pet Nutraceuticals/Supplements, Pet Treats, Pet Veterinary Diets) and by Distribution Channel (Convenience Stores, Online Channel, Specialty Stores, Supermarkets/Hypermarkets). The market volume and value are presented in metric ton and USD respectively. The key data points include the market size of pet food by products, distribution channels, and pets.

The number of pet cats adopted from shelters or rescues in the United States reached over *** million in 2023.

Over the last several decades, feral cats have moved from the fringes to the mainstream in animal welfare and sheltering. Although many best practice guidelines have been published by national non-profits and veterinary bodies, little is known about how groups “in the trenches” actually operate. Our study sought to address that gap through an online survey of feral cat care and advocacy organizations based in the United States. Advertised as “The State of the Mewnion,” its topics included a range of issues spanning non-profit administration, public health, caretaking and trapping, adoptions of friendly kittens and cats, veterinary medical procedures and policies, data collection and program efficacy metrics, research engagement and interest, and relationships with wildlife advocates and animal control agencies. Respondents from 567 organizations participated, making this the largest and most comprehensive study on this topic to date. Respondents came primarily from grassroots organizations. A majority reported no paid employees (74.6%), served 499 or fewer feral cats per year (75.0%), engaged between 1 and 9 active volunteers (54.9%), and did not operate a brick and mortar facility (63.7%). Some of our findings demonstrate a shared community of practice, including the common use of a minimum weight of 2.0 pounds for spay/neuter eligibility, left side ear tip removals to indicate sterilization, recovery holding times after surgery commonly reported as 1 night for male cats and 1 or 2 nights for females, requiring or recommending to adopters of socialized kittens/cats that they be kept indoor-only, and less than a quarter still engaging in routine testing of cats for FIV and FeLV. Our survey also reveals areas for improvement, such as most organizations lacking a declared goal with a measurable value and a time frame, only sometimes scanning cats for microchips, and about a third not using a standardized injection site for vaccines. This study paints the clearest picture yet available of what constitutes the standard practices of organizations serving feral and community cats in the United States.

CaTS-Bench Dataset

  Setup

After downloading, run tar -xzvf CaTSBench.tar.gz

to untar the directory.

  Contents





  Train

train_data contains four subdirectories:

gt_captions: Ground-truth captions, generated with contextual and statistical metadata alongside the time series. metadata: Contextual and statistical metadata for each time series plots: One plot image per time series time series: The numerical time series values

train_data contains 16k total… See the full description on the dataset page: https://huggingface.co/datasets/neurips2025submission/CaTS-Bench.