Kannada-MNIST

Kannada-MNIST是一个专为卡纳达语设计的手写数字数据集，由65名来自印度班加罗尔的志愿者创建，旨在作为原始MNIST数据集的直接替代品。该数据集包含60000张训练图像和10000张测试图像，每张图像大小为28x28像素。创建过程涉及数据收集、预处理、校验和训练测试分割。Kannada-MNIST的应用领域包括机器学习模型的训练和测试，特别是在手写数字识别领域，旨在提高模型对卡纳达语数字的识别能力。

Kannada-MNIST is a handwritten digit dataset specifically designed for the Kannada language. It was created by 65 volunteers from Bangalore, India, with the aim of serving as a direct drop-in replacement for the original MNIST dataset. This dataset comprises 60,000 training images and 10,000 test images, each with a resolution of 28×28 pixels. The dataset construction involves data collection, preprocessing, validation, and training-test splitting. Its application scenarios include the training and testing of machine learning models, particularly in the field of handwritten digit recognition, with the objective of improving the model's recognition ability for Kannada numerals.

搜集汇总

背景与挑战

背景概述

Kannada-MNIST是一个用于Kannada语言的手写数字图像数据集，设计作为经典MNIST数据集的替代品。该数据集曾用于Kaggle竞赛，吸引了全球1214支队伍参与，并已发表相关学术论文（arXiv:1908.01242）。