CNN for Old Japanese Character Classification

Center for Open Data in Humanities (人文学オープンデータ共同利用センター) launched Japanese Classics Character Dataset (日本古典籍字形データセット) in November 2016 [1]. This is a large dataset of various hand-written characters from classical documents written in the Edo period. Since these characters were written in running style, most of them are hard to read today (even for Japanese people). This example shows how t-SNE and CNN can classify the data.
[1] Japanese Classics Character Dataset: http://codh.rois.ac.jp/char-shape/