Counting values in an array, storing in a hash then making an array of hashes?

Hello,

Apologies but i've run in to another problem while attempting to make a sudoku solver that uses basic genetic programming.

This time I'm trying to count how many times each number occures in each board (boards stored in @offspring) then store the results in a hash.

I think each hash is being overwritten because the keys are the same, so I'm attempting to store an array of hashes.

I think I'm close, but my logic is still a bit off, as the length of my array of hashes is 1 but I expect it to be 4, with each element containing a hash. 4 is the size of the test population I am using, so there are 4 boards in @offspring.

I could have probably done the same thing with an array, but I confused myself thinking about it :frowning:

My code so far:

def fitness_check
hashes = Array.new
count = Hash.new 0
#Adds another loop to ensure all elements of offspring are checked separetly.
i = 0
#loop through every row of offspring (first to fourth element)
@offspring[i.to_i].each do |r|
#loop through every column of every row
r.each do |c|
#Count the numbers in each column
count[c] += 1
hashes[i] = count
end

#Need to stop the values in count getting overwritten by the next element in @offspring
puts hashes.inspect
puts hashes.length
end

Thanks in advance for any suggestions.

Jen.

totals = Hash.new(0)

arr = [1, 1, 2, 2, 3, 3, 1, 3]

arr.each do |val|
  totals[val] += 1
end

p totals

--output:--
{1=>3, 2=>2, 3=>3}

···

--
Posted via http://www.ruby-forum.com/.

3)

totals = []

boards = [
   [
      [1,1,1],
      [1,1,1],
      [1,1,1]
   ],

   [
      [2,2,2],
      [2,2,2],
      [2,2,2]
   ]
]

board_totals = []

boards.each do |this_board|
  this_board_totals = Hash.new(0)

  this_board.each do |row|
    row.each do |val|
      this_board_totals[val] += 1
    end
  end

  board_totals << this_board_totals
end

p board_totals

--output:--
[{1=>9}, {2=>9}]

···

--
Posted via http://www.ruby-forum.com/.

I'm not 100% positive that this is what you want, but it looks
suspiciously like what 7stud posted so here we go:

def fitness_check
  # loop to ensure all elements of offspring are checked separetly.
  hashes = @offspring.map do |board|
    Hash.new(0).tap do |count|

      # loop through every row of offspring (first to fourth element)
      board.each do |r|
        # loop through every column of every row
        r.each do |c|
          # Count the numbers in each column
          count[c] += 1
        end
      end

      count
    end
  end

  # debug output
  puts hashes.length
  p hashes
end

Kind regards

robert

···

On Mon, Mar 21, 2011 at 9:33 PM, Jen <jen.bottom@gmail.com> wrote:

Hello,

Apologies but i've run in to another problem while attempting to make a
sudoku solver that uses basic genetic programming.

This time I'm trying to count how many times each number occures in each
board (boards stored in @offspring) then store the results in a hash.

I think each hash is being overwritten because the keys are the same, so I'm
attempting to store an array of hashes.

I think I'm close, but my logic is still a bit off, as the length of my
array of hashes is 1 but I expect it to be 4, with each element containing a
hash. 4 is the size of the test population I am using, so there are 4 boards
in @offspring.

I could have probably done the same thing with an array, but I confused
myself thinking about it :frowning:

My code so far:

def fitness_check
hashes = Array.new
count = Hash.new 0
#Adds another loop to ensure all elements of offspring are checked
separetly.
i = 0
#loop through every row of offspring (first to fourth element)
@offspring[i.to_i].each do |r|
#loop through every column of every row
r.each do |c|
#Count the numbers in each column
count[c] += 1
hashes[i] = count
end
end

#Need to stop the values in count getting overwritten by the next element in
@offspring
puts hashes.inspect
puts hashes.length
end

Thanks in advance for any suggestions.

--
remember.guy do |as, often| as.you_can - without end
http://blog.rubybestpractices.com/