Neural Network with Rust

This project implements a simple neural network in Rust to solve the XOR problem. The neural network is capable of training itself using the backpropagation algorithm.

Use the fallowing command to get the static result

cargo run

XOR Test Result:

cargo test xor_test_nn

//to display the result
cargo test xor_test_nn -- --nocapture

    Test XOR [0-0] -> 0.07447127026483495
    Test XOR [0-1] -> 0.9442157962753881
    Test XOR [1-0] -> 0.9427510736015517
    Test XOR [1-1] -> 0.047543707694473566

AND

cargo test and_test_nn


//to display the result
cargo test and_test_nn -- --nocapture

Question for And --> 0-0 : [
    0.006176246532957251,]
Question for And --> 0-1 : [
    0.02496336563462987, ]
Question for And --> 1-0 : [
    0.023030271958667296,]
Question for And --> 1-1 : [
    0.966386645374941,
]

NAND

cargo test nand_test_nn


//to display the result
cargo test nand_test_nn -- --nocapture

Question for NAnd --> 0-0 : [
    0.9938314990590297,]
Question for NAnd --> 0-1 : [
    0.9796065695507552,]
Question for NAnd --> 1-0 : [
    0.9777487513661111,]
Question for NAnd --> 1-1 : [
    0.02887171712103688,]