CIFAR-10 vs CIFAR-100
| Feature | CIFAR-10 | CIFAR-100 |
|---|---|---|
| Number of Classes | 10 | 100 |
| Class Labels | Airplane, Automobile, Bird, Cat, Deer, Dog, Frog, Horse, Ship, Truck | Apple, Aquarium Fish, Baby, Bear, Beaver, Bed, Bee, Beetle, Bicycle, Bottle, etc. (total 100 classes) |
| Number of Images | 60,000 (50,000 training + 10,000 test) | 60,000 (50,000 training + 10,000 test) |
| Image Dimensions | 32×32 pixels | 32×32 pixels |
| Color Channels | 3 (RGB) | 3 (RGB) |
| Data Format | 32x32x3 numpy arrays | 32x32x3 numpy arrays |
| Train/Test Split | 50,000 training images / 10,000 test images | 50,000 training images / 10,000 test images |
| Per-Class Samples | 6,000 images per class | 600 images per class |
| Dataset Size | ~163 MB | ~163 MB |
| Dataset Creator | Alex Krizhevsky, Vinod Nair, and Geoffrey Hinton | Alex Krizhevsky, Vinod Nair, and Geoffrey Hinton |
| Year of Release | 2009 | 2009 |
| Applications | Image classification, object recognition, machine learning benchmarks | Fine-grained image classification, object recognition, machine learning benchmarks |
CIFAR 100 Dataset
The CIFAR-100 dataset is a dataset that is widely used in the field of computer vision, serving as a foundational tool for developing and testing machine learning models. This article provides a detailed exploration of the CIFAR-100 dataset and loading process.
Table of Content
- What is the CIFAR-100 Dataset?
- Classes and Superclasses
- Role of the CIFAR-100 Dataset in Computer Vision
- How to Load CIFAR-100 Dataset in TensorFlow
- CIFAR-10 vs CIFAR-100
- Applications of CIFAR-100 Dataset
- FAQs on CIFAR-100 Dataset