# This test create a database to adjust the best algorithm
# to use on correlation matrix
require(File.expand_path(File.dirname(__FILE__)+'/../helpers_benchmark.rb'))
require 'statsample'
require 'benchmark'

def create_dataset(vars,cases) 
  ran=Distribution::Normal.rng
  ds=vars.times.inject({}) {|ac,v|
    ac["x#{v}"]=Statsample::Vector.new_numeric(cases) {ran.call}
  ac
  }.to_dataset
end

def prediction_pairwise(vars,cases)
	Statsample::Bivariate.prediction_pairwise(vars,cases) / 10
end
def prediction_optimized(vars,cases)
	Statsample::Bivariate.prediction_optimized(vars,cases) / 10
end



if !File.exists?("correlation_matrix.ds") or File.mtime(__FILE__) > File.mtime("correlation_matrix.ds")
reps=100 #number of repetitions
ds_sizes=[5,10,30,50,100,150,200,500,1000]
ds_vars=[3,4,5,10,20,30,40]
#ds_sizes=[5,10]
#ds_vars=[3,5,20]
rs=Statsample::Dataset.new(%w{cases vars time_optimized time_pairwise})

ds_sizes.each do |cases|
  ds_vars.each do |vars|
      ds=create_dataset(vars,cases)
      time_optimized= Benchmark.realtime do
        reps.times { 
        Statsample::Bivariate.correlation_matrix_optimized(ds) 
        ds.clear_gsl
        }
      end
      
      time_pairwise= Benchmark.realtime do
        reps.times { 
        Statsample::Bivariate.correlation_matrix_pairwise(ds)
        }
      end
      
      puts "Cases:#{cases}, vars:#{vars} -> opt:%0.3f (%0.3f) | pair: %0.3f (%0.3f)" % [time_optimized, prediction_optimized(vars,cases), time_pairwise, prediction_pairwise(vars,cases)]
      
      rs.add_case({'cases'=>cases,'vars'=>vars,'time_optimized'=>Math.sqrt(time_optimized*1000),'time_pairwise'=>Math.sqrt(time_pairwise*1000)})
    end
  end
  
else
  rs=Statsample.load("correlation_matrix.ds")
end


rs.fields.each {|f| rs[f].type=:numeric}

rs['c_v']=rs.collect {|row| row['cases']*row['vars']}

rs.update_valid_data
rs.save("correlation_matrix.ds")
Statsample::Excel.write(rs,"correlation_matrix.xls")



rb=ReportBuilder.new(:name=>"Correlation matrix analysis")

rb.add(Statsample::Regression.multiple(rs[['cases','vars','time_optimized','c_v']],'time_optimized', :digits=>6))
rb.add(Statsample::Regression.multiple(rs[['cases','vars','time_pairwise','c_v']],'time_pairwise', :digits=>6))


rb.save_html("correlation_matrix.html")