Reduced Version of my NN solution to RubyWarrior. Without support for using different layered NN's and extra comments removed. Includes two evolved genomes, put one in a file called 'genome' with player.rb in warrior folder

genome_epic

0.4208665093792384,0.8271085823481578,-0.08677969236421801,-1.3812424909479102,0.25662357967972205,0.895993960666072,0.5969217612341047,1.003827461932993,0.3535111488721142,1.626234122590899,-0.9807624824831435,0.8455667360510928,-0.6596045396559488,-0.32330752252947614,0.30962364095597517,-1.051266687042697,-0.8619251071880633,1.0131278329152824,0.7045901871883101,0.5102103240087068,-0.5795087495450799,-1.4718975639160914,1.110149168791025,0.4848770449182779,-0.131034115009259,-0.8994602986777726,1.379669139856652,-0.24300620776607484,-0.032104129747120846,0.21746064790006747,-1.1751996077694868,-0.4416596710613584,1.2838430768445492,-0.33854260905174827,-0.5552927940371294,0.19992797470487422,-1.2768902810663525,-3.223132063475557,2.6814521898725237,-1.9607168460213784,1.9180731391978023,0.6077600309107104,0.4123090743791793,0.8246054475587798,-0.9596703206415195,0.2646596987299211,-0.06851920283518687,0.9944253709871688,0.24848358116043623,0.7495855784944395,-1.108429896105167,1.3753052571268305,3.181034120558026,-1.2709155470015525,0.7253500256394799,2.4192587701666324,2.122847393625867,-0.7075864717716149,2.694571198462502,-0.23738793552525383,-2.6191541131659126,0.6470749529047221,-2.0785370070988187,0.21706440490961854,-1.0164747730432557,0.16792697963601755,0.5023039485374504,-1.6222578128267933,-1.6384362302042,0.3516923946976447,-0.08501410283203104,0.10333251322753634,0.9694370284329388,0.7053274460037942,-1.9104000853472534,1.347447081663761,-0.875632334774423,0.4292051447798868,-0.1650455988119438,-0.3090559012185816,0.8122849956890801,-0.43795253002994095,-0.6324174808071471,1.052703281088451,-1.1261157583338193,1.073884982118626,1.602349239517356,-1.5930163275608602,-0.5929434775878311,-0.08861092539187976,2.2426990558759963,-0.6078690447899587,-1.4689188415414924,0.8433812929082517,-0.09344770334669705,0.5036286454281341,-0.6455383167824994,-2.2115467726742777,0.9784517143150431,-1.5474794887298606,-1.0121117734936058,1.315809794779084,-0.038522836988336806,-0.1922363540190457,0.34537500635109775,-0.07035164026793084,0.1304654458666375,0.046499273727501345,-0.776155984209477,-0.37063833219461223,2.953491431451073,-0.05017712876532365,-0.43929723081370353,-0.9906992496151041,-1.9884220925888678,-2.869539337671243,-1.4158283863183865,0.7505673085798398,0.9765036522488568,0.2733308317062427,1.7150446139071789,-1.1283890568444024,-0.11792247213316376,2.4010116280638307,-0.4041747776518825,0.17187082634317985,2.9376700582278765,-0.5871316925721045,1.3928431558645926,-0.1618869375393336,0.5749040004267305,-0.4636279527831687,1.3626708417630131,0.711740301132716,-0.0659435614151952,0.28541086127064585,-1.649405424787955,-1.1626701947387152,0.8682689002089043,-0.8377851096785819,-1.65359097309008,-0.818010200816063,-0.6911964701717429,1.273553027323357

genome_normal

0.5802744208575772,-0.28574107165038476,-0.4662748785179248,-1.2171913093525357,0.5656330027974735,-0.5586257421359373,0.5084206048395205,1.003827461932993,0.709080278772143,1.6392156876576782,-1.5816557649098901,1.324602802894768,0.8260058048894272,0.7178511436920003,-0.5804795029026557,-0.6639538111439616,-1.0830547059169757,1.9874278830514984,1.060462572920094,0.043450244058545695,-0.9386108314118661,-0.20380921105848493,0.9387378967454564,1.2449728237120754,-0.22972591303958378,-0.24970922127807615,1.62529564108517,0.5417832369383961,-0.11528627896579335,1.4118376011513845,-0.7192935919079477,-0.3072746576279173,0.5584597818074516,-0.7422452538004506,-0.4634791859355173,0.004736003702502156,-0.522971915073979,-1.6080762596194578,1.2867003151679492,-1.251724631556766,1.2491658636125211,1.3107157100065336,0.1900789190385367,1.8111732526381408,-1.401736698163552,0.5447534774158844,-0.8207816593573641,1.4844152049571304,0.7437535038324593,1.392577856212811,-1.0991699952789844,0.8711055513346737,1.6172844865648934,0.8405928179966495,0.4943506198504791,1.4284968522597425,-0.12065215283619857,-1.4419055454769207,2.7846954519824108,-0.7774835582067521,-2.145877659337063,0.9779931808120903,-0.410640494708302,0.03333410784760027,-1.451123469025841,0.5979415584663418,0.4598369719701315,-1.2593196304976257,-0.35010422037185673,1.1346703795954287,-0.9273859981034838,-0.4201631565453021,1.0212588282106947,0.700940688812856,-1.8844466294018205,0.7347508702944914,-1.0349553738651185,-0.04385884711391752,-0.13241728359821703,0.42457537461219197,1.3247043094193365,-0.8305801586055288,-0.7672892639721568,1.6630102110931597,-1.4630564793478578,1.7414649188338496,0.2974684205486997,-0.5212079800825287,0.18517333251448853,0.411561910031378,1.0719799685548463,-1.7452448790752302,-0.9517129751400798,-0.08448533537832714,-1.2325858448108846,-0.034609688077519296,-1.1319323029736506,-2.504005551714977,-0.3791599946626758,-1.1627861021688033,-0.694483154179575,0.897682867193146,0.1689166248283508,0.1244239416555194,-0.03729333603100804,0.18323346496734283,-0.34400573706550336,-0.09399750371134646,-1.7745907738007982,-0.6104668588467705,2.7888692908831696,0.35413361313989,-1.6159956231111794,-1.2513571752160617,-1.121825686506059,-1.96465470421759,-1.8425058033652832,1.23775809354918,0.6562636799454941,-0.17649029415055761,1.6087244520620438,0.4469884419946354,0.6381523759405975,-0.15787817155644912,-0.6123560840087449,0.48782473306474006,3.4922828293755748,-1.048601918281,0.7836219082927285,-0.8332571162284346,0.6919216046299118,-0.6089382368131837,0.2010862248447225,-0.3640148193914603,-0.7425593566026867,1.3646274903635693,-0.26189256664438787,-1.7190295526618184,-0.4506178405825584,0.1614189724301064,-1.5381221619979772,-0.8047572832141019,-0.3636923895878802,-0.07888040239556415

player.rb

class Player
  def initialize
    genome = File.open("./genome", "r"){|f| f.readlines}.join.split(",").map{|s| s.to_f} #Read genome from file.
    @brain = Brain.new({:in => 16, :inner => 6, :out => 8}, genome)  #Initialize warriors brain (neural net)
  end

  def play_turn(warrior)
    @previous_health ||= 20 
    inputs = input_array_for(warrior) #Sense world and present as an array of inputs for NN
    action, impulse = @brain.act_on(inputs) #send inputs to neural network and interpret its output as :action and :impulse   
    begin #send 'impulse' and 'action' from brain to warrior.  done inside rescue as brain may request actions the body can't yet do, like rest! in the eariler levels.  
      warrior.send(*["#{action}!", (action.eql?(:rest) ? nil : impulse)].compact)
    rescue NoMethodError => e
      puts "body failed to understand brain! #{e.message}"
    end
    @previous_health = warrior.health if warrior.respond_to?(:health)
  end

  def input_array_for warrior
    dirs = [:left, :forward, :right, :backward] #directions in which things can be
    things = [:wall, :enemy, :captive]          #type of things there can be
    vis_scale = [0, 0.6, 0.3] #used to scale the values returned by :look.     
    if warrior.respond_to?(:feel)     
      can_look = warrior.respond_to?(:look)
      inputs = things.map do |thing|  #for each of the things
        dirs.map do |dir|             #in each of the directions
          v = (warrior.feel(dir).send("#{thing}?").eql?(true) ? 1 : 0) #test if that thing is there, returning 1 for true else 0
          if can_look                 #if warrior can also look
            look = warrior.look(dir)  #look in direction
            v = v + look.map{|l| (l.send("#{thing}?").eql?(true) ? 1 : 0) * vis_scale[look.index(l)] }.max #reduce to a single val from given 3 ie [0,1,1] => [0, 0.6, 0.3] => [0.6]
          end
          v
        end
      end
    else           
      inputs = [1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0] #inputs for empty corridor.
    end
    inputs << (warrior.respond_to?(:shoot!) ? 1 : 0)
    w_health = warrior.respond_to?(:health) ? warrior.health : 20
    inputs << (1 - 1.0/20 * w_health).round(2) 
    inputs << ((@previous_health > w_health) ? 1 : 0) #sense of health dropping
    inputs << 1 #representational bias. yeah, I should prob explain this!  its REALLY important!    
    inputs.flatten
  end 
end

class Brain
  def initialize nodes, genome
    p1 = (nodes[:in] * nodes[:inner]) #number of genes(weights) needed for first layer
    @network = NeuralNetwork.new([
      NeuralLayer.new(genome[0..p1-1].in_groups_of(nodes[:in])), #init 1st layer with weights from genome
      NeuralLayer.new(genome[p1..(genome.size-1)].in_groups_of(nodes[:inner])) #init 2nd layer with weights from genome
    ]) 
  end
  
  def act_on inputs
    output = @network.process(inputs) #send inputs to neural net and return result as array of values (nodes).     
    if output[0].abs > output[1].abs #moving forward or backwards
      impulse = (output[0] > 0) ? :forward : :backward
    else #moving left or right
      impulse = (output[1] > 0) ? :left : :right
    end   
    actions = [[:walk, output[2]], [:attack, output[3]], [:rest, output[4]], [:rescue, output[5]], [:pivot, output[6]], [:shoot, output[7]]]
    action = actions.max_by{|grp| grp.last}.first #order actions by largest output value and select
    impulse = :rest if action.eql?(:rest) #rest is the only non-directional action (this line is only needed by assertions in BootCamp, not core to solutions funtion)
    [action, impulse] #return the action and impulse
  end
end

class NeuralNetwork
  def initialize layers
    @layers = layers[0..2]
  end
  def process inputs
    output = inputs 
    @layers.each{ |layer| output = layer.process(output) } #<---This calls the neural network calculation step
    output
  end
end

class NeuralLayer
  def initialize weights 
    @weights = weights
    @nodes = {:output => @weights.size, :input => @weights.first.size}
  end
  def process inputs
    @nodes[:output].times.map{ |i| inputs.zip(@weights[i]).map{|d| d.product_with_activation}.sum.round(2) } #<---Main NN calculation step.  
  end
end

class Array
  def sum; self.inject{|i,j| i.to_f + j.to_f}; end
  def product; self.inject{|i,j| i.to_f * j.to_f}; end 
  def product_with_activation; Math.sin(product); end
  def in_groups_of n
    col = []
    self.each_slice(n){|slice| col << slice}
    return col
  end
end