Datasets:
task_categories:
- image-classification
size_categories:
- 10K<n<100K
license: cc0-1.0
Here is the image dataset I used to train my model for image classification on dog breeds. It is a modified version of Gerry's 70 Dog Breeds-Image Data Set (much appreciated).
In this dataset, I removed four breeds due to being wild dogs, such as coyotes. I also added seven common breeds to the data to complete the 20 most popular dog breeds according to the American Kennel Club. Furthermore, I completely revamped some of the existing labels, such as the Pit Bull (which had multiple kinds mixed into one label), replacing the images to narrow it down to the Staffordshire Bull Terrier, the most common type of Pit Bull. Additionally, I performed some image cleaning by reducing repeats and incorrect pictures. Finally, I renamed several labels to fit the more standardized versions of the dogs' names.
If you would like to learn more about how I automated the image acquisition process by creating a custom Google search, you can find it in my GitHub repositoryhere. Thanks!