c# - Simple accord.net machine learning example -

i’m new machine learning , new accord.net (i code c#).

i want create simple project @ simple time series of data oscillate, want accord.net learn , predict next value be.

this data (time series) should like:

x - y

1 - 1  2 - 2  3 - 3  4 - 2  5 - 1  6 - 2  7 - 3  8 - 2  9 - 1 

then want predict following:

x - y

10 - 2  11 - 3  12 - 2  13 - 1  14 - 2  15 - 3 

can guys me out examples on how solve it?

a simple way use accord id3 decision tree.

the trick work out inputs use - can't train on x - tree won't learn future values of x - can build features derived x (or previous values of y) useful.

normally problems - make each prediction based on features derived previous values of y (the thing being predicted) rather x. assumes can observe y sequentially between each prediction (you can't predict arbitary x) i'll stick question presented.

i had go @ building accord id3 decision tree solve problem below. used few different values of x % n features - hoping tree work out answer this. in fact if i'd added (x-1) % 4 feature in single level attribute - guess point more let tree find patterns.

and here code :

    // sequence y follows     int[] ysequence = new int[] { 1, 2, 3, 2 };      // generates correct y given x     int calcy(int x) => ysequence[(x - 1) % 4];      // generates inputs - few differnt mod of x     int[] calcinputs(int x) => new int[] { x % 2, x % 3, x % 4, x % 5, x % 6 };       // http://stackoverflow.com/questions/40573388/simple-accord-net-machine-learning-example     [testmethod]     public void accordid3teststackoverflowquestion2()     {         // build training data set         int numtrainingcases = 12;         int[][] inputs = new int[numtrainingcases][];         int[] outputs = new int[numtrainingcases];          console.writeline("\t\t\t\t x \t y");         (int x = 1; x <= numtrainingcases; x++)         {             int y = calcy(x);             inputs[x-1] = calcinputs(x);             outputs[x-1] = y;             console.writeline("trainingdata \t " +x+"\t "+y);         }          // define how many values each input can have         decisionvariable[] attributes =         {             new decisionvariable("mod2",2),             new decisionvariable("mod3",3),             new decisionvariable("mod4",4),             new decisionvariable("mod5",5),             new decisionvariable("mod6",6)         };          // define how many outputs (+1 because y doesn't use zero)         int classcount = outputs.max()+1;          // create tree         decisiontree tree = new decisiontree(attributes, classcount);          // create new instance of id3 algorithm         id3learning id3learning = new id3learning(tree);          // learn training instances! populates tree         id3learning.learn(inputs, outputs);          console.writeline();         // try predict cases werent in training data         (int x = numtrainingcases+1; x <= 2* numtrainingcases; x++)         {             int[] query = calcinputs(x);              int answer = tree.decide(query); // makes prediction              assert.areequal(calcy(x), answer); // check answer expected - ie tree got right             console.writeline("prediction \t\t " + x+"\t "+answer);         }     } 

this output produces :

                 x   y trainingdata     1   1 trainingdata     2   2 trainingdata     3   3 trainingdata     4   2 trainingdata     5   1 trainingdata     6   2 trainingdata     7   3 trainingdata     8   2 trainingdata     9   1 trainingdata     10  2 trainingdata     11  3 trainingdata     12  2  prediction       13  1 prediction       14  2 prediction       15  3 prediction       16  2 prediction       17  1 prediction       18  2 prediction       19  3 prediction       20  2 prediction       21  1 prediction       22  2 prediction       23  3 prediction       24  2 

hope helps.

edit : following comments, below example modified train on previous values of target (y) - rather features derived time index (x). means can't start training @ start of series - need history of previous values of y. in example started @ x=9 because keeps same sequence.

        // sequence y follows     int[] ysequence = new int[] { 1, 2, 3, 2 };      // generates correct y given x     int calcy(int x) => ysequence[(x - 1) % 4];      // generates inputs - few differnt mod of x     int[] calcinputs(int x) => new int[] { calcy(x-1), calcy(x-2), calcy(x-3), calcy(x-4), calcy(x - 5) };     //int[] calcinputs(int x) => new int[] { x % 2, x % 3, x % 4, x % 5, x % 6 };       // http://stackoverflow.com/questions/40573388/simple-accord-net-machine-learning-example     [testmethod]     public void accordid3testteststackoverflowquestion2()     {         // build training data set         int numtrainingcases = 12;         int starttrainingat = 9;         int[][] inputs = new int[numtrainingcases][];         int[] outputs = new int[numtrainingcases];          console.writeline("\t\t\t\t x \t y");         (int x = starttrainingat; x < numtrainingcases + starttrainingat; x++)         {             int y = calcy(x);             inputs[x- starttrainingat] = calcinputs(x);             outputs[x- starttrainingat] = y;             console.writeline("trainingdata \t " +x+"\t "+y);         }          // define how many values each input can have         decisionvariable[] attributes =         {             new decisionvariable("y-1",4),             new decisionvariable("y-2",4),             new decisionvariable("y-3",4),             new decisionvariable("y-4",4),             new decisionvariable("y-5",4)         };          // define how many outputs (+1 because y doesn't use zero)         int classcount = outputs.max()+1;          // create tree         decisiontree tree = new decisiontree(attributes, classcount);          // create new instance of id3 algorithm         id3learning id3learning = new id3learning(tree);          // learn training instances! populates tree         id3learning.learn(inputs, outputs);          console.writeline();         // try predict cases werent in training data         (int x = starttrainingat+numtrainingcases; x <= starttrainingat + 2 * numtrainingcases; x++)         {             int[] query = calcinputs(x);              int answer = tree.decide(query); // makes prediction              assert.areequal(calcy(x), answer); // check answer expected - ie tree got right             console.writeline("prediction \t\t " + x+"\t "+answer);         }     } 

you consider training on differences between previous values of y - work better absolute value of y not important relative change.