accord.net

what should I use for non-linear vector regression with C# in Accord ? Thanks (traininginputs double[][] and trainingoutput double[] NOT int[]) Accord.NET provides a Support Vector Machine learning algorithm for regression problems in the SequentialMinimalOptimizationRegression class. There is an example application for this topic in the sample application's wiki page . Here is an example on how to use it: // Example regression problem. Suppose we are trying // to model the following equation: f(x, y) = 2x + y double[][] inputs = // (x, y) { new double[] { 0, 1 }, // 2*0 + 1 = 1 new double[] {

I'm classifing users with a multiclass svm (one-against-on), 3 classes. In binary, I would be able to plot the distribution of the weight of each feature in the hyperplan equation for different training sets. In this case, I don't really need a PCA to see stability of the hyperplan and relative importance of the features (reudced centered btw). What would the alternative be in multiclass svm, as for each training set you have 3 classifiers and you choose one class according to the result of the three classifiers (what is it already ? the class that appears the maximum number of times or the

I have 6 int parameters ranging from 0 to 100 The total combination of the numbers are 100^6 and each combination gives a result ranging approx. from -10000 to 100000 or even more. Input data example: MySimulation (57, 78, 20, 10, 90, 50) = 300 <- Best Result MySimulation (50, 80, 10, 90, 35, 8) = 200 MySimulation (4, 55, 40, 99, 40, 50) = -50 <- Worst Result The higher the result the better the combination of numbers are, I already have the calculation which gives a result, I only need AI to find a better combination of numbers which gives a higher result. Output data example: 55, 70, 25, 15,

I’m new to machine learning and new to accord.net (I code C#). I want to create a simple project where I look at a simple time series of data that oscillate, then I want accord.net to learn it and predict what the next value will be. This is what the data (time series) should look like: X - Y 1 - 1 2 - 2 3 - 3 4 - 2 5 - 1 6 - 2 7 - 3 8 - 2 9 - 1 Then I want it to predict the following: X - Y 10 - 2 11 - 3 12 - 2 13 - 1 14 - 2 15 - 3 Can you guys help me out with some examples on how to solve it? A simple way to do this would be to use an Accord ID3 decision tree. The trick is to work out what

I am a beginner at machine learning and Accord.net. I see how I can train it and then use it, but in an enterprise setting, running this on a server, how do I persist the learnings of Accord .net. Do I need to save all training data and retrain every time the application is restarted? Ususally there are Load and Save methods for persistance and re-creation of state. At which specific model are you looking at? Import the Accord.IO namespace in your code, by adding using Accord.IO; at the top of your source file. This way, you will be able to use the Serializer 's Save methods in any framework