Merge pull request #33 from JuliaRobotics/hotfix/warnserrorsjl1

removing errors and warnings also needed downstream

src/BallTree01.jl

@@ -359,9 +359,9 @@ function makeBallTree(_pointsMatrix::Array{Float64,2}, _weights::Array{Float64,1
 end
 
 function printBallTree(bt::BallTree)
-    @show round(bt.centers,15);
-    @show round(bt.weights,15);
-    @show round(bt.ranges,15);
+    @show round.(bt.centers,dims=15);
+    @show round.(bt.weights,dims=15);
+    @show round.(bt.ranges,dims=15);
     @show bt.left_child;
     @show bt.right_child;
     @show bt.highest_leaf;

src/BallTreeDensity01.jl

@@ -311,10 +311,10 @@ end
 
 function printBallTree(bd::BallTreeDensity, rnd=15)
   printBallTree(bd.bt);
-  @show round(bd.means,rnd);
-  @show round(bd.bandwidth,rnd);
-  @show round(bd.bandwidthMin,rnd);
-  @show round(bd.bandwidthMax,rnd);
+  @show round.(bd.means,digits=rnd);
+  @show round.(bd.bandwidth,digits=rnd);
+  @show round.(bd.bandwidthMin,digits=rnd);
+  @show round.(bd.bandwidthMax,digits=rnd);
 
   nothing
 end

src/DualTree01.jl

@@ -394,7 +394,7 @@ function evalAvgLogL(bd1::BallTreeDensity, bd2::BallTreeDensity)
   W = getWeights(bd2)
   ind = findall(L.==0.0)
   ll = nothing
-  if sum(findall(W[ind])) > 0
+  if sum(findall(x->x!=0, W[ind])) > 0
     # println("evalAvgLogL -- in if")
     ll=-Inf
   else
@@ -595,24 +595,24 @@ end
 
 function getKDERangeLinspace(bd::BallTreeDensity; extend::Float64=0.1, N::Int=201)
   v = getKDERange(bd,extend=extend)
-  return range(v[1], stop=v[2], length=N) #linspace(v[1],v[2],N)
+  return range(v[1], stop=v[2], length=N)
 end
 
 function getKDEMax(p::BallTreeDensity;N=200)
   m = zeros(p.bt.dims)
   for i in 1:p.bt.dims
     mm = marginal(p,[i])
     rangeV = getKDERange(mm)
-    X = linspace(rangeV[1],rangeV[2],N)
+    X = range(rangeV[1],stop=rangeV[2],length=N)
     # yV = evaluateDualTree(mm,X)
     yV = mm(reshape(collect(X), 1, N)) # TODO should allow AbstractArray
-    m[i] = X[findfirst(yV,maximum(yV))]
+    m[i] = X[something(findfirst(isequal(maximum(yV)), yV), 0)] # findfirst(yV,maximum(yV))
   end
   return m
 end
 
 function getKDEMean(p::BallTreeDensity)
-  return vec(Statistics.mean(getPoints(p),2))
+  return vec(Statistics.mean(getPoints(p),dims=2))
 end
 function getKDEfit(p::BallTreeDensity; distribution=MvNormal)
   fit(distribution, getPoints(p))

src/KDE01.jl

@@ -39,9 +39,6 @@ function getPoints(bd::BallTreeDensity)
     return pts
 end
 
-function getType(bd::BallTreeDensity)
-    return 0
-end
 
 #s = zeros(dens.D,dens.N);
 #s(:,double(dens.perm(dens.N + (1:dens.N)))+1) = dens.bandwidth(:,dens.N + (1:dens.N));
@@ -80,7 +77,7 @@ function marginal(bd::BallTreeDensity, ind::Array{Int,1})
   p = kde!(pts[ind,:],sig[ind], wts)
 end
 
-function randKernel(N::Int, M::Int, t::Int)
+function randKernel(N::Int, M::Int, ::Type{KernelDensityEstimate.GaussianKer}) #t::Int)
   return randn(N,M)
 end
 

src/KernelDensityEstimate.jl

@@ -60,5 +60,4 @@ A Julia package for Kernel Density Estimation and approximations of their produc
 
 """
 
-
 end

src/MSGibbs01.jl

@@ -252,14 +252,14 @@ function printGlbs(g::GbGlb, tag=Union{})
     end
     println("Ndim=$(g.Ndim), Ndens=$(g.Ndens), Nlevels=$(g.Nlevels), dNp=$(g.dNp), dNpts=$(g.dNpts)")
     @show g.ind
-    @show round(g.particles,2)
-    @show round(g.variance,2)
-    @show round(g.p,2)
-    @show round(g.Malmost,2)
-    @show round(g.Calmost,2)
+    @show round.(g.particles, digits=2)
+    @show round.(g.variance, digits=2)
+    @show round.(g.p, digits=2)
+    @show round.(g.Malmost, digits=2)
+    @show round.(g.Calmost, digits=2)
     @show g.levelList
     @show g.levelListNew
-    @show round(g.newPoints,4)
+    @show round.(g.newPoints, digits=4)
     @show g.newIndices
 end