CVJan 29, 2017
The HASYv2 dataset
arXiv:1701.08380v137 citations
Originality Synthesis-oriented
AI Analysis
This provides a new dataset for researchers in machine learning, particularly for tasks similar to MNIST, but it is incremental as it builds upon existing dataset concepts.
The paper introduces the HASYv2 dataset, a publicly available collection of 168,233 instances across 369 classes of single symbols, designed for classification and verification challenges with pre-defined folds for cross-validation.
This paper describes the HASYv2 dataset. HASY is a publicly available, free of charge dataset of single symbols similar to MNIST. It contains 168233 instances of 369 classes. HASY contains two challenges: A classification challenge with 10 pre-defined folds for 10-fold cross-validation and a verification challenge.