yes, spot on. And that character recognition software will be …. a JMP neural network. The data was sourced from the MNIST database (http://yann.lecun.com/exdb/mnist) – a popular dataset for machine learning.
As currently presented, the pixel grid I am using to render the images needs to be transposed – hence the reason that the squiggles are not immediately recognisable as characters.
As a postscript, anyone disappointed that they are not hieroglyphs – this book is a great intro: “How To Read Egyptian Hieroglyphs”, Collier & Manley, published by the British Museum Press”.
Looks like a test for character recognition software
yes, spot on. And that character recognition software will be …. a JMP neural network. The data was sourced from the MNIST database (http://yann.lecun.com/exdb/mnist) – a popular dataset for machine learning.
As currently presented, the pixel grid I am using to render the images needs to be transposed – hence the reason that the squiggles are not immediately recognisable as characters.
As a postscript, anyone disappointed that they are not hieroglyphs – this book is a great intro: “How To Read Egyptian Hieroglyphs”, Collier & Manley, published by the British Museum Press”.