Neural Network 5-Input Parity

---

---

Project Description

---

Created using Python with Pycharm as an IDE

This program utilizes a neural network to tell a user whether a 5 bit string contains an even or odd number of 1s. It takes as input a file containing a training set of 5 bit strings along with whether those strings have an even or odd number of 1s. Once trained, it will output what it thinks the answer is for every possible 5 bit binary number.

The program was a small team project primarily completed in a pair programming fashion.

---

The Network

---

The neural network uses an input layer, a hidden layer, and an output layer along with forward and back propagation to train it. When a training item is received, it uses the weights it currently has(and these are random to start) and will make a guess. Then, if the training set says that the network was incorrect, it will adjust the weights connecting the various layers. This procedure continues until the network reaches a high enough accuracy level, at which point it will take every possibly combination of 1s and 0s in a 5 bit string and output to the user what it thinks the answers are, along with the confidence.

---

Important Note about Code

---

If you are interested in the code for this particular project, please contact me at any of the available contacts I have posted here. The code itself is private to help maintain the integrity of the course at UVU. Thank you for your understanding.