Viterbi Algorithm on Wikipedia written in ruby

The Viterbi algorithm can solve the most likely state sequence problem of the Hidden Markov Model.

On Wikipedia, a python example is provided for the Viterbi algorithm. The original Article address is as follows:

Http://zh.wikipedia.org/wiki/%E7%BB%B4%E7%89%B9%E6%AF%94%E7% AE %97%E6%B3%95

Since we are learning Ruby recently, we have migrated this algorithm from Python to Ruby. the syntax of these two languages is very close, so it is not difficult to move forward, I hope you can understand the basic theory of the Viterbi algorithm before using the code.

 

 

##
#encoding: utf-8
puts ‘This is Viterbi’
$ states = [: Healthy,: Fever]
#puts "length: # {$ states.length}"
$ obervastions = [: normal,: cold,: dizzy]

$ start_probability = {Healthy: 0.6, Fever: 0.4}
# puts $ start_probability [$ states [0]]

$ transition_probability = {
    Healthy: {Healthy: 0.7, Fever: 0.3},
    Fever: {Healthy: 0.4, Fever: 0.6},
}

$ emission_probability = {
    Healthy: {normal: 0.5, cold: 0.4, dizzy: 0.1},
    Fever: {normal: 0.1, cold: 0.3, dizzy: 0.6}
}

def print_dptable (v)
  puts ‘‘
  for i in 0..v.length
    puts "% 7d"% i
  end

  for y in v [0] .keys
    puts "% 5s"% y
    for t in 0 ... v.length
      puts "% .7s"% ("% f"% v [t] [y])
    end
  end
end


def viterbi (obs, states, start_p, trans_p, emit_p)
  v = [{}]
  path = {}

# Initialize base cases (t == 0)
  states.each {| y |
    v [0] [y] = start_p [y] * emit_p [y] [obs [0]]
    path [y] = [y]
  }

# Run Viterbi for t> 0
  for t in 1 ... obs.length
    v << {}
    newpath = {}

    for y in states
      
      prob, state = states.map {| y0 |
      [v [t-1] [y0] * trans_p [y0] [y] * emit_p [y] [obs [t]], y0]} .max
      v [t] [y] = prob
      newpath [y] = path [state] + [y]
    end

    # Do n‘t need to remember the old paths
    path = newpath
  end

  print_dptable v
  prob, state = states.map {| y | [v [obs.size-1] [y], y]} .max
  return prob, path [state]
end

def example
  viterbi $ obervastions,
          $ states,
          $ start_probability,
          $ transition_probability,
          $ emission_probability
end

puts example