Add MLPClassifier + help example

release-packaging/Classes/FluidMLP.sc

@@ -1,5 +1,11 @@
 FluidBaseMLP : FluidDataClient {
-		*new {|server, hidden = #[3,3] , activation = 0, maxIter = 100, learnRate = 0.0001, momentum = 0.9, batchSize = 50, validation = 0.2|
+	const <identity = 0;
+	const <sigmoid  =  1;
+	const <relu = 2;
+	const <tanh = 3;
+
+
+	*new {|server, hidden = #[3,3] , activation = 0, maxIter = 100, learnRate = 0.0001, momentum = 0.9, batchSize = 50, validation = 0.2|
 		var hiddenCtrlLabels;
 		hidden = [hidden.size]++hidden;
 
@@ -19,12 +25,6 @@ FluidBaseMLP : FluidDataClient {
 }
 
 FluidMLPRegressor : FluidBaseMLP {
-
-  const <identity = 0;
-  const <sigmoid  =  1;
-  const <relu = 2;
-  const <tanh = 3;
-
 	*new {|server, hidden = #[3,3] , activation = 0, maxIter = 100, learnRate = 0.0001, momentum = 0.9, batchSize = 50, validation = 0.2|
 		^super.new(server,hidden,activation, maxIter,learnRate, momentum, batchSize,validation)
 	}
@@ -47,3 +47,27 @@ FluidMLPRegressor : FluidBaseMLP {
       [sourceBuffer.asUGenInput, targetBuffer.asUGenInput, layer], action);
 	}
 }
+
+FluidMLPClassifier : FluidBaseMLP {
+	*new {|server, hidden = #[3,3] , activation = 0, maxIter = 100, learnRate = 0.0001, momentum = 0.9, batchSize = 50, validation = 0.2|
+		^super.new(server,hidden,activation, maxIter,learnRate, momentum, batchSize,validation)
+	}
+
+	fit{|sourceDataSet, targetLabelSet, action|
+	   this.prSendMsg(\fit,
+			[sourceDataSet.asSymbol, targetLabelSet.asSymbol],
+			action,numbers(FluidMessageResponse,_,1,_)
+		);
+	}
+
+	predict{ |sourceDataSet, targetLabelSet, action|
+		this.prSendMsg(\predict,
+			[sourceDataSet.asSymbol, targetLabelSet.asSymbol],
+			action);
+	}
+
+	predictPoint { |sourceBuffer, action|
+		this.prSendMsg(\predictPoint,
+			[sourceBuffer.asUGenInput], action,string(FluidMessageResponse,_,_));
+	}
+}

release-packaging/HelpSource/Classes/FluidMLPClassifier.schelp

@@ -0,0 +1,152 @@
+TITLE:: FluidMLPClassifier
+summary:: Classification with a neural network
+categories:: Undocumented classes
+related:: Classes/FluidMLPRegressor, Classes/FluidDataSet
+
+DESCRIPTION::
+
+
+
+CLASSMETHODS::
+
+METHOD:: new
+(describe method here)
+
+ARGUMENT:: server
+(describe argument here)
+
+ARGUMENT:: hidden
+(describe argument here)
+
+ARGUMENT:: activation
+(describe argument here)
+
+ARGUMENT:: maxIter
+(describe argument here)
+
+ARGUMENT:: learnRate
+(describe argument here)
+
+ARGUMENT:: momentum
+(describe argument here)
+
+ARGUMENT:: batchSize
+(describe argument here)
+
+ARGUMENT:: validation
+(describe argument here)
+
+returns:: (describe returnvalue here)
+
+
+INSTANCEMETHODS::
+
+METHOD:: predictPoint
+(describe method here)
+
+ARGUMENT:: sourceBuffer
+(describe argument here)
+
+ARGUMENT:: action
+(describe argument here)
+
+returns:: (describe returnvalue here)
+
+METHOD:: fit
+(describe method here)
+
+ARGUMENT:: sourceDataSet
+(describe argument here)
+
+ARGUMENT:: targetLabelSet
+(describe argument here)
+
+ARGUMENT:: action
+(describe argument here)
+
+returns:: (describe returnvalue here)
+
+METHOD:: predict
+(describe method here)
+
+ARGUMENT:: sourceDataSet
+(describe argument here)
+
+ARGUMENT:: targetLabelSet
+(describe argument here)
+
+ARGUMENT:: action
+(describe argument here)
+
+returns:: (describe returnvalue here)
+
+
+EXAMPLES::
+
+code::
+(
+~classifier = FluidMLPClassifier(s,hidden:[6],validation:0,momentum:0.1,learnRate:0.01);
+~sourcedata= FluidDataSet(s,\mlpclassify_help_examples);
+~labels = FluidLabelSet(s,\mlpclassify_help_labels);
+~testdata = FluidDataSet(s,\mlpclassify_help_test);
+~predictedlabels = FluidLabelSet(s,\mlpclassify_help_mapping);
+)
+//Make some clumped 2D points and place into a DataSet
+(
+~centroids = [[0.5,0.5],[-0.5,0.5],[0.5,-0.5],[-0.5,-0.5]];
+~categories = [\red,\orange,\green,\blue];
+~trainingset = Dictionary();
+~labeldata = Dictionary();
+4.do{ |i|
+	64.do{ |j|
+		~trainingset.put("mlpclass"++i++\_++j, ~centroids[i].collect{|x| x.gauss(0.5/3)});
+		~labeldata.put("mlpclass"++i++\_++j,[~categories[i]]);
+	}
+};
+~sourcedata.load(Dictionary.with(*[\cols->2,\data->~trainingset]));
+~labels.load(Dictionary.with(*[\cols->1,\data->~labeldata]));
+)
+
+//Fit the classifier to the example DataSet and labels, and then run prediction on the test data into our mapping label set
+~classifier.fit(~sourcedata,~labels,action:{|loss| ("Trained"+loss).postln});
+
+//make some test data
+(
+~testset = Dictionary();
+4.do{ |i|
+	64.do{ |j|
+		~testset.put("mlpclass_test"++i++\_++j, ~centroids[i].collect{|x| x.gauss(0.5/3)});
+	}
+};
+~testdata.load(Dictionary.with(*[\cols->2,\data->~testset]));
+)
+
+//Run the test data through the network, into the predicted labelset
+~classifier.predict(~testdata,~predictedlabels,action:{"Test complete".postln});
+
+//get labels from server
+~predictedlabels.dump(action:{|d| ~labelsdict = d["data"]});
+//Visualise: we're hoping to see colours neatly mapped to quandrants...
+(
+c = Dictionary();
+c.add("red"->Color.red);
+c.add("blue"->Color.blue);
+c.add("green"->Color.green);
+c.add("orange"->Color.new255(255, 127, 0));
+e = 200 * ((~centroids + 1) * 0.5).flatten(1).unlace;
+w = Window("scatter", Rect(128, 64, 200, 200));
+w.drawFunc = {
+    Pen.use {
+		~testset.keysValuesDo{|k,v|
+			var x = v[0].linlin(-1,1,200,0).asInteger;
+			var y = v[1].linlin(-1,1,200,0).asInteger;
+			var r = Rect(x,y,5,5);
+			Pen.fillColor = c.at(~labelsdict[k][0]);
+			Pen.fillOval(r);
+		}
+    }
+};
+w.refresh;
+w.front;
+)
+::

src/FluidManipulation/FluidManipulation.cpp

@@ -13,6 +13,7 @@
 #include <clients/nrt/PCAClient.hpp>
 #include <clients/nrt/MDSClient.hpp>
 #include <clients/nrt/MLPRegressorClient.hpp>
+#include <clients/nrt/MLPClassifierClient.hpp>
 #include <clients/rt/AudioTransportClient.hpp>
 #include <clients/rt/FluidDataSetWr.hpp>
 #include <FluidSCWrapper.hpp>
@@ -38,4 +39,5 @@ PluginLoad(FluidSTFTUGen)
   makeSCWrapper<NRTThreadedAudioTransportClient>("FluidBufAudioTransport",ft);
   makeSCWrapper<NRTThreadedDataSetWriter>("FluidDataSetWr", ft);
   makeSCWrapper<NRTThreadedMLPRegressorClient>("FluidMLPRegressor",ft);
+  makeSCWrapper<NRTThreadedMLPClassifierClient>("FluidMLPClassifier",ft);
 }