A PyQt GUI app that uses Back-Propagation Neural Networks to train and match user defined patterns.
Basically the app tests 10 7 segment display pattern with their actual values like the following. A 7 segment display looks like the following.
---0---
| |
1 2
| |
---3---
| |
4 5
| |
---6---
A 7bit-input vs 4bit-output map would look like this.
[
# 0 1 2 3 4 5 6
[[1, 1, 1, 0, 1, 1, 1], [0, 0, 0, 0]], # 0
[[0, 0, 1, 0, 0, 1, 0], [0, 0, 0, 1]], # 1
[[1, 0, 1, 1, 1, 0, 1], [0, 0, 1, 0]], # 2
[[1, 0, 1, 1, 0, 1, 1], [0, 0, 1, 1]], # 3
[[0, 1, 1, 1, 0, 1, 0], [0, 1, 0, 0]], # 4
[[1, 1, 0, 1, 0, 1, 1], [0, 1, 0, 1]], # 5
[[1, 1, 0, 1, 1, 1, 1], [0, 1, 1, 0]], # 6
[[1, 1, 1, 0, 0, 1, 0], [0, 1, 1, 1]], # 7
[[1, 1, 1, 1, 1, 1, 1], [1, 0, 0, 0]], # 8
[[1, 1, 1, 1, 0, 1, 1], [1, 0, 0, 1]] # 9
]The program trains a Neural Network with 7bit input, 7bit hidden and 4bit output nodes. Then the user can click and draw any number. On clicking Check, the user provided segment values will be tested and a predicted value from the network will be shown.
To run, execute python3 Widget.py.
Then all the input patterns will be trained.
Creating Neural Network...
Training patterns...
error 10.45545 iterations 0
error 3.79253 iterations 1000
error 2.73814 iterations 2000
error 3.66140 iterations 3000
.............................
Training complete!
And finally the user can test any custom pattern and get predicted results.
Written by Neil Schemenauer <nas@arctrix.com>