CHAPTER_4

HOME  /  LESSONS  /  CHAPTER_4

Chapter_4
An Artificial Neural Network for Recognizing English Letters

Step 1. Run Program4Z_1.exe


Step 2. Use the ‘Convergence Rate’ button (Button ①) on the screen to set the rate value to 0.001. Each time you click this button, a different value will appear.


Step 3. Click ’Learn’ button (Button ②) to start the training process of the program. The image shown above is the screen that changes while the program is running. As the learning progresses, it is programmed that the output value of the artificial neural network is displayed from the left column to the right. “A’ is displayed for 1.0, “B’ for 2.0, and both ‘A’ and ‘B’ for 2.5.


Step 4. Check how the output value changes and RMS error of the artificial neural network in the ‘Python execution screen’ which is hidden below the above screen. The following shows a part of the ‘Python execution screen’.

Results


As the number of iterations increases to 100 and over 100, one can see that the letters in the first row are getting closer to letter ‘A’ and the letters in the last row are getting closer to ‘H’.

KamelConsulting, LLC.

CEO. Byung S. Moon | Tel. +82-10-6627-4306 | byung.s.moon@gmail.com ㅣ Biz License 314-86-26086

2F, DICIA ,20 Daedeok-daero 512, Daejeon, S. Korea


2022 Kamel Consulting, LLC