Artificial Neural Networks with Java: Tools for Building Neural Network Applications

© The Author(s), under exclusive license to APress Media, LLC, part of Springer Nature 2022

I. LivshinArtificial Neural Networks with Javahttps://doi.org/10.1007/978-1-4842-7368-5_14

14. Classification of Handwritten Digits

Igor Livshin¹  

(1)

Chicago, IL, USA

 

In the previous chapter, we prepared two input files.

• The input file for network training:

  C:\Image_Recognition\MNIS_Data\ mnist_train_input_Norm.csv

• The input file for network testing:

  C:\Image_Recognition\MNIS_Data\ mnist_test_input_Norm.csv

Just to remind you, the format of the records in both files is as follows: The first 784 fields contain the value of the grayscale pixels’ intensity (24*24 = 784 fields). These are the input fields for the network. The next ten fields of the record indicate ...