scores.rb

scores.rb

#!/usr/bin/env ruby

# do the statistics of renaissance results
#
# probably should choose gem 'enumerable-statistics' in the future.
# so far, we only need simple procedures.
#
def median(a)
  copy = a.sort
  return copy[a.size / 2]
end

def mean(a)
  return a.inject(0.0, :+) / a.size
end

# standard deviation
# treat 'a' is a sample, so use N-1 correction
def stddev(a)
  avg = mean(a)
  sum = a.inject(0.0) { |sum, e| sum + (e - avg) ** 2 }
  variance = sum / (a.size - 1)
  std = Math.sqrt(variance)
end

WARMUP_ITER = 3

def renassance_result(output)
  results = []
  output.each do |line|
    # ====== chi-square (apache-spark) [default], iteration 57 completed (795.509 ms) ======
    line.match /iteration \d+ completed \((\d+\.\d+) ms\)/ do |m|
      #puts line, m[1]
      results << m[1].to_f
    end
  end

  #puts "warmup #{WARMUP_ITER} iterations"
  # drop first WARMUP_ITER results
  results.shift WARMUP_ITER
  printf "means = %.3f, stddev = %.3f, median = %.3f\n", mean(results), stddev(results), median(results)
end

if $0 == __FILE__
  output = ARGF.readlines
  renassance_result output
end