Training Datasets
In machine learning, neural networks are trained on datasets to optimize the model to recognize patterns and make predictions across different domains.
Common datasets
The following datasets are used for benchmarking and research purposes.
- MNIST: 70.000 images of handwritten digits (0-9)
- CIFAR-10: 60.000 32x32 color images of 10 classes (planes, cars, birds, etc.)
- ImageNet: 14m images labeled across thousands of categories.
- Fashion-MNIST: 70.000 gray-scale images of clothing items (10 categories)
- COCO (Common Objects in Context): 330.000 images with annotations
- UCI Machine Learning Repository: Collection of datasets for tasks