pokemon_name_generator 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: a69c8787f7aae6ede03ceacb61be806b2ca42b10949657df5d1ddb4aba026d5b
+  data.tar.gz: b3dfe791b03f31d986f9abe4d6ab4fdb66c4bb102afaa14ebb55ddf5f121c314
+SHA512:
+  metadata.gz: 0ae71b4218fd57a08cefe4717ebbff446007cf80bb15e1d877595035b32776f55a33553d1bc31609f94d9465d6770a12d122f71dffd3fdcf69e46779e63c7153
+  data.tar.gz: 2fd38d0192ff55b2c5a2cf13bec54491ece8a9eb3c5cd1b9019a8628a5358833deb150731d2eca55fa2934198de36eebe9443e55b4212d1fa9ad3d4eb6b1d05b

data/README.md

@@ -0,0 +1,44 @@
+# Pokémon Name Generator
+
+A command line ruby utility to generate fake (and sometimes accidentally real)
+Pokémon names.
+
+## Usage
+
+Usage can be very simple.
+
+```bash
+$ bin/pokeng generate
+```
+
+You can choose between algorithms.
+
+```bash
+$ bin/pokeng generate --algorithm=naive
+$ bin/pokeng generate --algorithm=markov
+```
+
+Some algorithms have additional configuration options.
+
+```bash
+$ bin/pokeng generate --algorithm=markov --context=2
+```
+
+You can also produce a number of names at once.
+Useful for exporting to other apps.
+
+```bash
+$ bin/pokeng generate --algorithm=markov --context=3 --number=100000 > generated_names.txt
+```
+
+The default options are chosen to produce the most realistic names.
+
+## Testing the Algorithm
+
+You can put an algorithm through a testbed to find out how well it performs.
+It counts how many times an actual Pokémon name is produced (from the test and
+training data seperately) and how often a overly long or short name is produced.
+
+```bash
+$ bin/pokeng test --algorithm=markov --context=2 --number=500000
+```

data/bin/pokeng

@@ -0,0 +1,7 @@
+#!/usr/bin/env ruby
+
+$LOAD_PATH.unshift(File.expand_path('../lib', __dir__))
+
+require 'pokemon_name_generator'
+
+Dry::CLI.new(PokemonNameGenerator::CLI::Commands).call

data/lib/pokemon_name_generator/cli/commands.rb

@@ -0,0 +1,88 @@
+require "bundler/setup"
+require "dry/cli"
+
+module PokemonNameGenerator
+  module CLI
+    module Commands
+      extend Dry::CLI::Registry
+
+      class Version < Dry::CLI::Command
+        desc "Print version"
+
+        def call(*)
+          puts PokemonNameGenerator::VERSION
+        end
+      end
+
+      class Generate < Dry::CLI::Command
+        desc "Generate one or more Pokémon Names"
+
+        option :algorithm, default: "markov", values: %w[naive markov], desc: "Algorithm to use"
+
+        option :context, type: :integer, default: 3, desc: "Number of links in the Markov Chain"
+
+        option :number, type: :integer, default: 1, desc: "Number of names to generate"
+
+        def call(**options)
+          corpus = Corpus.new
+
+          algorithm = case options.fetch(:algorithm)
+          when "naive" then Naive.new(corpus.pokemon_phonemes)
+          when "markov" then Markov.new(corpus.pokemon_phonemes, context_length: options.fetch(:context).to_i)
+          end
+
+          options.fetch(:number).to_i.times { puts algorithm.generate_name }
+        end
+      end
+
+      class Test < Dry::CLI::Command
+        desc "Test an algorithm"
+
+        option :algorithm, required: true, values: %w[naive markov], desc: "Algorithm to test"
+
+        option :context, required: true, type: :integer, desc: "Number of links in the Markov Chain"
+
+        option :number, type: :integer, default: 10_000, desc: "Number of names to generate"
+
+        def call(**options)
+          corpus = Corpus.new
+          actual_pokemon = corpus.pokemon
+          all_data = corpus.pokemon_phonemes.shuffle
+
+          longest_name = actual_pokemon.max_by(&:length)
+          shortest_name = actual_pokemon.min_by(&:length)
+
+          mid_point = (all_data.size + 1) / 2
+          training_data = all_data[..mid_point]
+          test_data = all_data[mid_point..]
+
+          algorithm = case options.fetch(:algorithm)
+          when "naive" then Naive.new(training_data)
+          when "markov" then Markov.new(training_data, context_length: options.fetch(:context).to_i)
+          end
+
+          puts "============================="
+          puts "🧪 Generator: #{algorithm.name}"
+          puts "============================="
+
+          generated_names = options.fetch(:number).to_i.times.map { algorithm.generate_name }
+
+          puts "Training Data: #{training_data.size}"
+          puts "Test Data: #{test_data.size}"
+          puts "Generated Names: #{generated_names.size}"
+          puts "Unique Generated Names: #{generated_names.uniq.size}"
+          puts "Training Data Names Generated: #{training_data.map { |datum| datum.join("") }.intersection(generated_names.uniq).size}"
+          puts "Test Data Names Generated: #{test_data.map { |datum| datum.join("") }.intersection(generated_names.uniq).size}"
+          puts "Test Data Names Generated: #{test_data.map { |datum| datum.join("") }.intersection(generated_names.uniq)}"
+          puts "Overly Short Names: #{generated_names.count { |name| name.size < shortest_name.size }}"
+          puts "Overly Long Names: #{generated_names.count { |name| name.size > longest_name.size }}"
+          puts ""
+        end
+      end
+
+      register "version", Version, aliases: ["-v", "--version"]
+      register "generate", Generate, aliases: ["g", "generate"]
+      register "test", Test, aliases: ["t", "test"]
+    end
+  end
+end

data/lib/pokemon_name_generator/cli.rb

@@ -0,0 +1,5 @@
+module PokemonNameGenerator
+  module CLI
+    require "pokemon_name_generator/cli/commands"
+  end
+end

data/lib/pokemon_name_generator/corpus.rb

@@ -0,0 +1,46 @@
+require_relative "./phoneme"
+require_relative "./pokemon"
+
+class Corpus
+  def pokemon_letters
+    @pokemon_letters ||= load_pokemon_letters.values
+  end
+
+  def pokemon_phonemes
+    @pokemon_phonemes ||= load_pokemon_phonemes.values
+  end
+
+  def phonemes
+    @phonemes ||= Phoneme.load_data
+  end
+
+  def pokemon
+    @pokemon ||= Pokemon.load_data
+  end
+
+  private
+
+  # This generates a hash so we can debug it easily
+  def load_pokemon_letters
+    pokemon.map { |name| [name, name.chars] }.to_h
+  end
+
+  # This generates a hash so we can debug it easily
+  def load_pokemon_phonemes
+    pokemon.map do |this_pokemon_name|
+      remaining_pokemon_name = this_pokemon_name
+      this_pokemon_phonemes = []
+
+      loop do
+        break if remaining_pokemon_name.empty?
+
+        phoneme = phonemes.find { |phoneme| remaining_pokemon_name.start_with?(phoneme) }
+
+        this_pokemon_phonemes << phoneme
+        remaining_pokemon_name = remaining_pokemon_name.sub(phoneme, "")
+      end
+
+      [this_pokemon_name, this_pokemon_phonemes]
+    end.to_h
+  end
+end

data/lib/pokemon_name_generator/markov.rb

@@ -0,0 +1,64 @@
+class Markov
+  def initialize(training_data, context_length:)
+    @training_data = training_data
+    @context_length = context_length
+  end
+
+  def name
+    "Markov[#{@context_length}]"
+  end
+
+  def generate_name
+    pokemon_name = ""
+    context = []
+    current_phoneme = statistics[context].sample
+
+    loop do
+      break if current_phoneme.nil?
+
+      pokemon_name << current_phoneme
+      context << current_phoneme
+      context = context.drop(1) if context.size > context_length
+      current_phoneme = statistics[context].sample
+    end
+
+    pokemon_name
+  end
+
+  private
+
+  attr_reader :training_data, :context_length
+
+  def statistics
+    @statistics ||= load_statistics
+  end
+
+  def load_statistics
+    markov_statistics = {}
+
+    training_data.each do |this_pokemon_phonemes|
+      context = []
+
+      this_pokemon_phonemes.each do |this_pokemon_phoneme|
+        if markov_statistics[context]
+          markov_statistics[context] << this_pokemon_phoneme
+        else
+          markov_statistics[context] = [this_pokemon_phoneme]
+        end
+
+        context = [context, this_pokemon_phoneme].flatten
+        context = context.drop(1) if context.size > context_length
+      end
+
+      if markov_statistics[context]
+        markov_statistics[context] << nil
+      else
+        markov_statistics[context] = [nil]
+      end
+    end
+
+    markov_statistics[[]] = markov_statistics[[]].uniq
+
+    markov_statistics
+  end
+end

data/lib/pokemon_name_generator/naive.rb

@@ -0,0 +1,28 @@
+class Naive
+  def initialize(training_data)
+    @training_data = training_data
+  end
+
+  def name
+    "Naïve"
+  end
+
+  def generate_name
+    statistics[:phoneme_count_distribution].sample.times.map { statistics[:phoneme_distribution].sample }.join
+  end
+
+  private
+
+  attr_reader :training_data, :context_length
+
+  def statistics
+    @statistics ||= load_statistics
+  end
+
+  def load_statistics
+    {
+      phoneme_count_distribution: training_data.map(&:count),
+      phoneme_distribution: training_data.flatten
+    }
+  end
+end

data/lib/pokemon_name_generator/phoneme.rb

@@ -0,0 +1,7 @@
+module Phoneme
+  DATA = "./data/phonemes.txt"
+
+  def self.load_data
+    File.readlines(DATA, chomp: true)
+  end
+end

data/lib/pokemon_name_generator/pokemon.rb

@@ -0,0 +1,10 @@
+module Pokemon
+  DATA = "./data/pokemon.txt"
+
+  def self.load_data
+    File.readlines(DATA, chomp: true)
+      .map { |line| line[5..] } # drop the number
+      .map(&:downcase)
+      .reject { |name| name =~ /[♀♂.'2\-é:\ ]/ } # don't deal with special characters yet
+  end
+end

data/lib/pokemon_name_generator/version.rb

@@ -0,0 +1,3 @@
+module PokemonNameGenerator
+  VERSION = "0.1.0"
+end

data/lib/pokemon_name_generator.rb

@@ -0,0 +1,9 @@
+module PokemonNameGenerator
+  require "pokemon_name_generator/cli"
+  require "pokemon_name_generator/corpus"
+  require "pokemon_name_generator/markov"
+  require "pokemon_name_generator/naive"
+  require "pokemon_name_generator/phoneme"
+  require "pokemon_name_generator/pokemon"
+  require "pokemon_name_generator/version"
+end

metadata

@@ -0,0 +1,54 @@
+--- !ruby/object:Gem::Specification
+name: pokemon_name_generator
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Tony Rowan
+autorequire:
+bindir: bin
+cert_chain: []
+date: 2022-07-14 00:00:00.000000000 Z
+dependencies: []
+description:
+email: trowan812@gmail.com
+executables:
+- pokeng
+extensions: []
+extra_rdoc_files: []
+files:
+- README.md
+- bin/pokeng
+- lib/pokemon_name_generator.rb
+- lib/pokemon_name_generator/cli.rb
+- lib/pokemon_name_generator/cli/commands.rb
+- lib/pokemon_name_generator/corpus.rb
+- lib/pokemon_name_generator/markov.rb
+- lib/pokemon_name_generator/naive.rb
+- lib/pokemon_name_generator/phoneme.rb
+- lib/pokemon_name_generator/pokemon.rb
+- lib/pokemon_name_generator/version.rb
+homepage: https://github.com/tony-rowan/pokemon-name-generator
+licenses:
+- MIT
+metadata: {}
+post_install_message:
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.3.7
+signing_key:
+specification_version: 4
+summary: A tool to create new Pokémon names
+test_files: []