Real correct for verbose

Author: theogf

test/test_BSVM.jl

@@ -49,7 +49,7 @@ ps = []; t_full = 0; t_sparse = 0; t_stoch =0;
 #### FULL MODEL EVALUATION ####
 if fullm
     println("Testing the full batch model")
-    t_full = @elapsed fullmodel = OMGP.BatchBSVM(X,y,Autotuning=autotuning,noise=noise,kernel=kernel,verbose=3)
+    t_full = @elapsed fullmodel = OMGP.BatchBSVM(X,y,Autotuning=autotuning,noise=noise,kernel=kernel,verbose=verbose)
     t_full += @elapsed fullmodel.train(iterations=20)
     y_full = fullmodel.predictproba(X_test); acc_full = 1-sum(abs.(sign.(y_full.-0.5)-y_test))/(2*length(y_test))
     if doPlots
@@ -61,7 +61,7 @@ end
 # #### SPARSE MODEL EVALUATION ####
 if sparsem
     println("Testing the sparse model")
-    t_sparse = @elapsed sparsemodel = OMGP.SparseBSVM(X,y,Stochastic=false,Autotuning=autotuning,verbose=3,m=20,noise=noise,kernel=kernel)
+    t_sparse = @elapsed sparsemodel = OMGP.SparseBSVM(X,y,Stochastic=false,Autotuning=autotuning,verbose=verbose,m=20,noise=noise,kernel=kernel)
     t_sparse += @elapsed sparsemodel.train(iterations=100)
     y_sparse = sparsemodel.predictproba(X_test); acc_sparse = 1-sum(abs.(sign.(y_sparse.-0.5)-y_test))/(2*length(y_test))
     if doPlots
@@ -73,7 +73,7 @@ end
 # #### STOCH. MODEL EVALUATION ####
 if stochm
     println("Testing the sparse stochastic model")
-    t_stoch = @elapsed stochmodel = OMGP.SparseBSVM(X,y,Stochastic=true,batchsize=10,Autotuning=autotuning,verbose=3,m=20,noise=noise,kernel=kernel)
+    t_stoch = @elapsed stochmodel = OMGP.SparseBSVM(X,y,Stochastic=true,batchsize=10,Autotuning=autotuning,verbose=verbose,m=20,noise=noise,kernel=kernel)
     t_stoch += @elapsed stochmodel.train(iterations=1000)
     y_stoch = stochmodel.predictproba(X_test); acc_stoch = 1-sum(abs.(sign.(y_stoch.-0.5)-y_test))/(2*length(y_test))
     if doPlots

test/test_Regression.jl

@@ -33,7 +33,7 @@ kernel = RBFKernel(2.0)
 autotuning=false
 
 println("Testing the full model")
-t_full = @elapsed fullmodel = OMGP.GPRegression(X,y,noise=noise,Autotuning=autotuning,kernel=kernel,verbose=3)
+t_full = @elapsed fullmodel = OMGP.GPRegression(X,y,noise=noise,Autotuning=autotuning,kernel=kernel,verbose=verbose)
 t_full += @elapsed fullmodel.train()
 y_full = fullmodel.predict(X_test); rmse_full = norm(y_full-y_test,2)/sqrt(length(y_test))
 if doPlots
@@ -42,7 +42,7 @@ if doPlots
 end
 
 println("Testing the sparse model")
-t_sparse = @elapsed sparsemodel = OMGP.SparseGPRegression(X,y,Stochastic=false,Autotuning=autotuning,verbose=3,m=20,noise=noise,kernel=kernel)
+t_sparse = @elapsed sparsemodel = OMGP.SparseGPRegression(X,y,Stochastic=false,Autotuning=autotuning,verbose=verbose,m=20,noise=noise,kernel=kernel)
 t_sparse += @elapsed sparsemodel.train(iterations=100)
 y_sparse = sparsemodel.predict(X_test); rmse_sparse = norm(y_sparse-y_test,2)/sqrt(length(y_test))
 if doPlots
@@ -52,7 +52,7 @@ if doPlots
 end
 
 println("Testing the sparse stochastic model")
-t_stoch = @elapsed stochmodel = OMGP.SparseGPRegression(X,y,Stochastic=true,batchsize=20,Autotuning=autotuning,verbose=2,m=20,noise=noise,kernel=kernel)
+t_stoch = @elapsed stochmodel = OMGP.SparseGPRegression(X,y,Stochastic=true,batchsize=20,Autotuning=autotuning,verbose=verbose,m=20,noise=noise,kernel=kernel)
 t_stoch += @elapsed stochmodel.train(iterations=200)
 y_stoch = stochmodel.predict(X_test); rmse_stoch = norm(y_stoch-y_test,2)/sqrt(length(y_test))
 if doPlots

test/test_StudentT.jl

@@ -40,7 +40,7 @@ stochm = true
 
 if fullm
     println("Testing the full model")
-    t_full = @elapsed global fullmodel = OMGP.BatchStudentT(X,y,noise=noise,kernel=kernel,verbose=3,Autotuning=autotuning,ν=ν)
+    t_full = @elapsed global fullmodel = OMGP.BatchStudentT(X,y,noise=noise,kernel=kernel,verbose=verbose,Autotuning=autotuning,ν=ν)
     t_full += @elapsed fullmodel.train(iterations=100)
     y_full = fullmodel.predict(X_test); rmse_full = norm(y_full-y_test,2)/sqrt(length(y_test))
     if doPlots
@@ -51,7 +51,7 @@ end
 
 if sparsem
     println("Testing the sparse model")
-    t_sparse = @elapsed global sparsemodel = OMGP.SparseStudentT(X,y,Stochastic=false,Autotuning=autotuning,verbose=3,m=m,noise=noise,kernel=kernel,ν=ν)
+    t_sparse = @elapsed global sparsemodel = OMGP.SparseStudentT(X,y,Stochastic=false,Autotuning=autotuning,verbose=verbose,m=m,noise=noise,kernel=kernel,ν=ν)
     t_sparse += @elapsed sparsemodel.train(iterations=1000)
     y_sparse = sparsemodel.predict(X_test); rmse_sparse = norm(y_sparse-y_test,2)/sqrt(length(y_test))
     if doPlots
@@ -63,7 +63,7 @@ end
 
 if stochm
     println("Testing the sparse stochastic model")
-    t_stoch = @elapsed stochmodel = OMGP.SparseStudentT(X,y,Stochastic=true,batchsize=20,Autotuning=autotuning,verbose=2,m=m,noise=noise,kernel=kernel,ν=ν)
+    t_stoch = @elapsed stochmodel = OMGP.SparseStudentT(X,y,Stochastic=true,batchsize=20,Autotuning=autotuning,verbose=verbose,m=m,noise=noise,kernel=kernel,ν=ν)
     t_stoch += @elapsed stochmodel.train(iterations=1000)
     y_stoch = stochmodel.predict(X_test); rmse_stoch = norm(y_stoch-y_test,2)/sqrt(length(y_test))
     if doPlots

test/test_XGPC.jl

@@ -56,7 +56,7 @@ ps = []; t_full = 0; t_sparse = 0; t_stoch = 0;
 # # #### FULL MODEL EVALUATION ####
 if fullm
     println("Testing the full model")
-    t_full = @elapsed fullmodel = OMGP.BatchXGPC(X,y,noise=noise,kernel=kernel,verbose=3,Autotuning=true)
+    t_full = @elapsed fullmodel = OMGP.BatchXGPC(X,y,noise=noise,kernel=kernel,verbose=verbose,Autotuning=true)
     t_full += @elapsed fullmodel.train(iterations=20)
     y_full = fullmodel.predictproba(X_test); acc_full = 1-sum(abs.(sign.(y_full.-0.5)-y_test))/(2*length(y_test))
     if doPlots
@@ -67,7 +67,7 @@ end
 # # #### SPARSE MODEL EVALUATION ####
 if sparsem
     println("Testing the sparse model")
-    t_sparse = @elapsed sparsemodel = OMGP.SparseXGPC(X,y,Stochastic=false,Autotuning=true,ϵ=1e-6,verbose=3,m=N_indpoints,noise=1e-10,kernel=kernel,OptimizeIndPoints=false)
+    t_sparse = @elapsed sparsemodel = OMGP.SparseXGPC(X,y,Stochastic=false,Autotuning=true,ϵ=1e-6,verbose=verbose,m=N_indpoints,noise=1e-10,kernel=kernel,OptimizeIndPoints=false)
     metrics,savelog = OMGP.getLog(sparsemodel,X_test=X_test,y_test=y_test)
     t_sparse += @elapsed sparsemodel.train(iterations=100)#,callback=savelog)
     y_sparse = sparsemodel.predictproba(X_test); acc_sparse = 1-sum(abs.(sign.(y_sparse.-0.5)-y_test))/(2*length(y_test))
@@ -81,7 +81,7 @@ end
 #### STOCH. SPARSE MODEL EVALUATION ###.
 if ssparsem
     println("Testing the sparse stochastic model")
-    t_stoch = @elapsed stochmodel = OMGP.SparseXGPC(X,y,Stochastic=true,batchsize=40,Autotuning=true,verbose=2,m=N_indpoints,noise=noise,kernel=kernel,OptimizeIndPoints=false)
+    t_stoch = @elapsed stochmodel = OMGP.SparseXGPC(X,y,Stochastic=true,batchsize=40,Autotuning=true,verbose=verbose,m=N_indpoints,noise=noise,kernel=kernel,OptimizeIndPoints=false)
     metrics,savelog = OMGP.getLog(stochmodel,X_test=X_test,y_test=y_test)
     t_stoch += @elapsed stochmodel.train(iterations=1000)#,callback=savelog)
     y_stoch = stochmodel.predictproba(X_test); acc_stoch = 1-sum(abs.(sign.(y_stoch.-0.5)-y_test))/(2*length(y_test))