tinn/README.md

Tinn (Tiny Neural Network) is a 200 line dependency free neural network library written in C99.

#include "Tinn.h"
#include <stdio.h>

#define SETS 4
#define NIPS 2
#define NHID 8
#define NOPS 1
#define ITER 2000
#define RATE 1.0f

int main()
{
    // This example learns XOR.
    float in[SETS][NIPS] = {
        { 0, 0 },
        { 0, 1 },
        { 1, 0 },
        { 1, 1 },
    };
    float tg[SETS][NOPS] = {
        { 0 },
        { 1 },
        { 1 },
        { 0 },
    };
    // Build.
    const Tinn tinn = xtbuild(NIPS, NHID, NOPS);
    // Train.
    for(int i = 0; i < ITER; i++)
    {
        float error = 0.0f;
        for(int j = 0; j < SETS; j++)
            error += xttrain(tinn, in[j], tg[j], RATE);
        printf("%.12f\n", error / SETS);
    }
    // Predict.
    for(int i = 0; i < SETS; i++)
    {
        const float* pd = xtpredict(tinn, in[i]);
        printf("%f :: %f\n", tg[i][0], (double) pd[0]);
    }
    // Cleanup.
    xtfree(tinn);
    return 0;
}

For a more complicated demo on how to learn hand written digits, get some training data:

wget http://archive.ics.uci.edu/ml/machine-learning-databases/semeion/semeion.data

And if you're on Linux / MacOS just build and run Tinn with the test file:

make; ./tinn

If you're on Windows it's:

mingw32-make & tinn.exe

For the layman not accustomed to makefiles the makefile devolves into:

gcc test.c Tinn.c -lm

The training data consists of hand written digits written both slowly and quickly. Each line in the data set corresponds to one handwritten digit. Each digit is 16x16 pixels in size giving 256 inputs to the neural network.

At the end of the line 10 digits signify the hand written digit:

0: 1 0 0 0 0 0 0 0 0 0
1: 0 1 0 0 0 0 0 0 0 0
2: 0 0 1 0 0 0 0 0 0 0
3: 0 0 0 1 0 0 0 0 0 0
4: 0 0 0 0 1 0 0 0 0 0
...
9: 0 0 0 0 0 0 0 0 0 1

This gives 10 outputs to the neural network. The test program will output the accuracy for each digit. Expect above 99% accuracy for the correct digit, and less that 0.1% accuracy for the other digits.

Tips

• Tinn will never use more than the C standard library.

• Tinn is great for embedded systems. Train a model on your powerful desktop and load it onto a microcontroller and use the analog to digital converter to predict real time events.

• The Tinn source code will always be less than 200 lines. Functions externed in the Tinn header are protected with the xt namespace standing for externed tinn.

• Tinn can easily be multi-threaded with a bit of ingenuity but the master branch will remain single threaded to aid development for embedded systems.

• Tinn does not seed the random number generator. Do not forget to do so yourself.

• Always shuffle your input data. Shuffle again after every training iteration.

• Get greater training accuracy by annealing your learning rate. For instance, multiply your learning rate by 0.99 every training iteration. This will zero in on a good learning minima.

Disclaimer

Tinn is not a fully featured neural network C library like Kann, or Genann:

https://github.com/attractivechaos/kann

https://github.com/codeplea/genann

Tinn is minimal and easy to digest.