Have you seen the movie, Rain Man? Do you remember the scene when Dustin Hoffman can count the exact number of toothpicks on the floor in the blink of an eye. This scene gave the idea to implement it as a machine learning example. We don't count toothpicks but dots in an image.
For further details see the blog post: https://bhaxor.blog.hu/2019/03/10/the_semantic_mnist
The sources can be found at https://gitlab.com/nbatfai/smnist
Here is an introductory video: https://youtu.be/-tSRwJgVpJk
The experiments called SMNIST for Humans are intended to measure the capacity of the parallel individuation system in humans.
The experiments called SMNIST for Machines are similar to previous ones but they investigate computer programs.
Source: smnistg.cpp
Source: smnistg-no-centering.cpp
Source: smnistg-disjunct-test-set-no-centering.cpp
Source: smnistg-disjunct-1px-test-set-no-centering.cpp
Source: smnistg-hard-disjunct-test-set-no-centering.cpp
Source: smnistg-hard-1px-disjunct-test-set-no-centering.cpp
Source: smnistg-s2-disjunct.cpp
Source: smnistg-s2-hard.cpp
Source: smnistg-s2-disjunct-pow102x+.cpp
Source: smnistg-s2-hard-pow102x+.cpp
Source: smnistg-sX5-hard-pow10x+.cpp
Source: smnistg-s5-hard-pow10x+.cpp
Bugfix: 1544 of 60000 train images contain points that have the same coords
"1px": In the former datasets, dots are 3x3 pixels, now they are 1 pixel.
The case of 0 dots is handled standalone.
Last modified: 13 June 2019
Norbert Bátfai, PhD
batfai.norbert@inf.unideb.hu
https://arato.inf.unideb.hu/batfai.norbert/
IT Dept, UD