update README based on 0.1.9 mods, formatting, bump version to 0.1.10

README.md

@@ -44,51 +44,59 @@ assert_approximate_equality_of_nested_list net.weights, net2.weights, 0.00000000
 
 ## Roadmap
 
-- [ ]  Add RNN
-
-- [ ]  Port from `ai4r`:
-  - [ ]  classifiers
-    - [ ]  classifier.rb
-    - [ ]  hyperpipes.rb
-    - [ ]  ib1.rb
-    - [ ]  id3.rb
-    - [ ]  multilayer_perceptron.rb
-    - [ ]  naive_bayes.rb
-    - [ ]  one_r.rb
-    - [ ]  prism.rb
-    - [ ]  simple_linear_regression.rb
-    - [ ]  votes.rb
-      - [ ]  zero_r.rb
-  - [ ]  clusterers
-    - [ ]  average_linkage.rb
-    - [ ]  bisecting_k_means.rb
-    - [ ]  centroid_linkage.rb
-    - [ ]  clusterer.rb
-    - [ ]  complete_linkage.rb
-    - [ ]  diana.rb
-    - [ ]  k_means.rb
-    - [ ]  median_linkage.rb
-    - [ ]  single_linkage.rb
-    - [ ]  ward_linkage_hierarchical.rb
-    - [ ]  ward_linkage.rb
-      - [ ]  weighted_average_linkage.rb
-  - [ ]  data
-    - [ ]  data_set.rb
-    - [ ]  parameterizable.rb
-    - [ ]  proximity.rb
-      - [ ]  statistics.rb
-  - [ ]  experiment
-      - [ ]  classifier_evaluator.rb
-  - [ ]  genetic_algorithm
-      - [ ]  genetic_algorithm.rb
-  - [ ]  neural_network
+- [x] Generate an error history plot using `AsciiBarCharter` and `error_distance_history` , e.g.:
+```
+plot: '▇▊▂_▅▅▅_▅_▅▅▅▅_▅▅__▅_▅____▅___'
+```
+(Run `crystal spec spec_examples` to see more error history plot examples. NOTE: These run short training sessions, so some tests are likely to fail some of the time.)
+
+- [ ] Add Cmn ("Connectable Mini Networks") (WIP)
+
+- [ ] Add RNN
+
+- [ ] Port from `ai4r`:
+  - [ ] classifiers
+    - [ ] classifier.rb
+    - [ ] hyperpipes.rb
+    - [ ] ib1.rb
+    - [ ] id3.rb
+    - [ ] multilayer_perceptron.rb
+    - [ ] naive_bayes.rb
+    - [ ] one_r.rb
+    - [ ] prism.rb
+    - [ ] simple_linear_regression.rb
+    - [ ] votes.rb
+      - [ ] zero_r.rb
+  - [ ] clusterers
+    - [ ] average_linkage.rb
+    - [ ] bisecting_k_means.rb
+    - [ ] centroid_linkage.rb
+    - [ ] clusterer.rb
+    - [ ] complete_linkage.rb
+    - [ ] diana.rb
+    - [ ] k_means.rb
+    - [ ] median_linkage.rb
+    - [ ] single_linkage.rb
+    - [ ] ward_linkage_hierarchical.rb
+    - [ ] ward_linkage.rb
+      - [ ] weighted_average_linkage.rb
+  - [ ] data
+    - [ ] data_set.rb
+    - [ ] parameterizable.rb
+    - [ ] proximity.rb
+      - [ ] statistics.rb
+  - [ ] experiment
+      - [ ] classifier_evaluator.rb
+  - [ ] genetic_algorithm
+      - [ ] genetic_algorithm.rb
+  - [ ] neural_network
     - [x] backpropagation.rb
-    - [ ]  hopfield.rb
-  - [ ]  som
-    - [ ]  layer.rb
-    - [ ]  node.rb
-    - [ ]  som.rb
-    - [ ]  two_phase_layer.rb
+    - [ ] hopfield.rb
+  - [ ] som
+    - [ ] layer.rb
+    - [ ] node.rb
+    - [ ] som.rb
+    - [ ] two_phase_layer.rb
 
 If you'd like another class of Ai4r ported, feel free to submit a [new issue](https://github.com/drhuffman12/ai4cr/issues/new).
 

shard.yml

@@ -1,5 +1,5 @@
 name: ai4cr
-version: 0.1.9
+version: 0.1.10
 
 authors:
   - Daniel Huffman <drhuffman12@yahoo.com>

spec/ai4cr/neural_network/cmn/connected_net_set/sequencial_spec.cr

@@ -11,14 +11,14 @@ describe Ai4cr::NeuralNetwork::Cmn::ConnectedNetSet::Sequencial do
       [-0.4, 0.9, -0.4, -0.7],
       [0.1, 0.8, 0.9, -0.0],
       [-0.7, -0.3, -0.6, -0.7],
-      [1.0, 0.2, 0.6, -0.5]
+      [1.0, 0.2, 0.6, -0.5],
     ]
     hard_coded_weights1 = [
       [-0.4, 0.8],
       [-1.0, -0.3],
       [-0.6, 0.6],
       [0.2, -0.3],
-      [1.0, -0.1]
+      [1.0, -0.1],
     ]
 
     puts "hard_coded_weights0: #{hard_coded_weights0.each { |a| puts a.join("\t") }}"
@@ -28,29 +28,26 @@ describe Ai4cr::NeuralNetwork::Cmn::ConnectedNetSet::Sequencial do
     expected_outputs_guessed_after = [0.454759979898907, 0.635915600435646]
 
     it "the 'outputs_guessed' start as zeros" do
-
       net0 = Ai4cr::NeuralNetwork::Cmn::MiniNet::Exp.new(height: 3, width: 4)
       net1 = Ai4cr::NeuralNetwork::Cmn::MiniNet::Exp.new(height: 4, width: 2)
       cns = Ai4cr::NeuralNetwork::Cmn::ConnectedNetSet::Sequencial(Ai4cr::NeuralNetwork::Cmn::MiniNet::Exp).new([net0, net1])
-  
+
       puts "net0.weights: #{net0.weights.map { |a| a.map { |b| b.round(1) } }}"
       puts "net1.weights: #{net1.weights.map { |a| a.map { |b| b.round(1) } }}"
-  
+
       net0.init_network
       net0.learning_rate = 0.25
       net0.momentum = 0.1
       net0.weights = hard_coded_weights0.clone
       # puts "\nnet0 (BEFORE): #{net0.to_json}\n"
 
-
       net1.init_network
       net1.learning_rate = 0.25
       net1.momentum = 0.1
       net1.weights = hard_coded_weights1.clone
       # puts "\nnet1 (BEFORE): #{net1.to_json}\n"
 
       puts "\ncns (BEFORE): #{cns.to_json}\n"
-
 
       outputs_guessed_before = net1.outputs_guessed.clone
 
@@ -61,7 +58,7 @@ describe Ai4cr::NeuralNetwork::Cmn::ConnectedNetSet::Sequencial do
       net0 = Ai4cr::NeuralNetwork::Cmn::MiniNet::Exp.new(height: 3, width: 4)
       net1 = Ai4cr::NeuralNetwork::Cmn::MiniNet::Exp.new(height: 4, width: 2)
       cns = Ai4cr::NeuralNetwork::Cmn::ConnectedNetSet::Sequencial(Ai4cr::NeuralNetwork::Cmn::MiniNet::Exp).new([net0, net1])
-  
+
       puts "net0.weights: #{net0.weights.map { |a| a.map { |b| b.round(1) } }}"
       puts "net1.weights: #{net1.weights.map { |a| a.map { |b| b.round(1) } }}"
 
@@ -71,15 +68,14 @@ describe Ai4cr::NeuralNetwork::Cmn::ConnectedNetSet::Sequencial do
       net0.weights = hard_coded_weights0.clone
       # puts "\nnet0 (BEFORE): #{net0.to_json}\n"
 
-
       net1.init_network
       net1.learning_rate = 0.25
       net1.momentum = 0.1
       net1.weights = hard_coded_weights1.clone
       # puts "\nnet1 (BEFORE): #{net1.to_json}\n"
 
       puts "\ncns (BEFORE): #{cns.to_json}\n"
-      
+
       outputs_guessed_before = cns.net_set.last.outputs_guessed.clone
 
       cns.eval(inputs)
@@ -88,6 +84,5 @@ describe Ai4cr::NeuralNetwork::Cmn::ConnectedNetSet::Sequencial do
 
       assert_approximate_equality_of_nested_list outputs_guessed_after, expected_outputs_guessed_after
     end
-
   end
 end

spec_examples/ai4cr/neural_network/backpropagation_spec.cr

@@ -8,7 +8,6 @@ require "json"
 describe Ai4cr::NeuralNetwork::Backpropagation do
   describe "#train" do
     describe "with a shape of [256,3]" do
-
       describe "using image data (input) and shape flags (output) for triangle, square, and cross" do
         correct_count = 0
 
@@ -170,7 +169,6 @@ describe Ai4cr::NeuralNetwork::Backpropagation do
     end
 
     describe "with a shape of [256,300,3]" do
-
       describe "using image data (input) and shape flags (output) for triangle, square, and cross" do
         correct_count = 0
 

spec_examples/ai4cr/neural_network/cmn/connected_net_set/sequencial_spec.cr

@@ -39,7 +39,7 @@ describe Ai4cr::NeuralNetwork::Cmn::ConnectedNetSet::Sequencial do
         net0 = Ai4cr::NeuralNetwork::Cmn::MiniNet::Exp.new(height: 256, width: 300, error_distance_history_max: 60)
         net1 = Ai4cr::NeuralNetwork::Cmn::MiniNet::Exp.new(height: 300, width: 3, error_distance_history_max: 60)
         cns = Ai4cr::NeuralNetwork::Cmn::ConnectedNetSet::Sequencial(Ai4cr::NeuralNetwork::Cmn::MiniNet::Exp).new([net0, net1])
-    
+
         # net.learning_rate = rand
         qty = 500
         qty_10_percent = qty // 10

src/ai4cr/neural_network/backpropagation.cr

@@ -219,7 +219,7 @@ module Ai4cr
         outputs = outputs.map { |v| v.to_f }
         eval(inputs)
         load_expected_outputs(outputs)
-        backpropagate # (outputs)
+        backpropagate   # (outputs)
         calculate_error # (outputs)
       end
 
@@ -387,7 +387,7 @@ module Ai4cr
       def load_expected_outputs(expected_outputs)
         @expected_outputs.map_with_index! { |v, i| expected_outputs[i] }
       end
-      
+
       # Calculate quadratic error for a expected output value
       # Error = 0.5 * sum( (expected_value[i] - output_value[i])**2 )
       def calculate_error # (expected_outputs)

src/ai4cr/neural_network/cmn/connected_net_set/sequencial.cr

@@ -12,11 +12,10 @@ module Ai4cr
           property net_set
 
           def initialize(@net_set : Array(T))
-
           end
 
           def validate!
-            index_max = @net_set.size-1
+            index_max = @net_set.size - 1
             width_height_mismatch = [0..index_max].map do |index|
               @net_set[index].width != @net_set[index + 1].height
             end.any?
@@ -43,18 +42,17 @@ module Ai4cr
 
           # TODO: utilize until_min_avg_error
           def train(inputs_given, outputs_expected, until_min_avg_error = 0.1)
-
             @net_set.each_with_index do |net, index|
               index == 0 ? net.step_load_inputs(inputs_given) : net.step_load_inputs(@net_set[index - 1].outputs_guessed)
               net.step_calc_forward
             end
 
-            index_max = @net_set.size-1
+            index_max = @net_set.size - 1
             (0..index_max).to_a.reverse.each do |index|
               net = @net_set[index]
 
-              index == index_max ? net.step_load_outputs(outputs_expected) : net.step_load_outputs(@net_set[index + 1].input_deltas[0..@net_set[index + 1].height-1])
-              
+              index == index_max ? net.step_load_outputs(outputs_expected) : net.step_load_outputs(@net_set[index + 1].input_deltas[0..@net_set[index + 1].height - 1])
+
               net.step_calculate_error
               net.step_backpropagate
             end

src/ai4cr/version.cr

@@ -1,3 +1,3 @@
 module Ai4cr
-  VERSION = "0.1.9"
+  VERSION = "0.1.10"
 end