ruby_linear_regression 0.0.2 → 0.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 7fdd3245c2d76d40b4e6a1992644360e0cab3d74
-  data.tar.gz: 3c02457473a0695ac8eae4803a8b04799e8ba4b0
+  metadata.gz: cfebab034391389c2a317a1ab52e38657c7fa02b
+  data.tar.gz: 2846a47266a55f14e0c9ee42a61d36a9c40ac882
 SHA512:
-  metadata.gz: bea431b65d547919c489aa894d34a20a4d70b41733c9eedc32fcd6447d8feb4cd951c084f12167281e88a442d99ba799b969437a732327f7fc5ee132eb4954bf
-  data.tar.gz: e5242100d48114b33da5ec3e729f2b554df7258a7738583440233d85030b00a7771e7946b27f55a9b58a725e29655c3659a27ef80264bec66c49e4fbc88a24e7
+  metadata.gz: 88db7a42087da42fcc74cfd8363625269aace47110e7383a2b78add3756ee7599b648331186e583502d506eaa698f2e67006cda949217bbb19ba01dfe748dc7b
+  data.tar.gz: cbd925a7eb8f9a1f721c80c37ededd062f9ec885f353ab1425101ec481c268c534efa35958b7692321c3b96fc82dbd506b7093590ff7d987b6b3e5b9c6ac35aa

data/lib/ruby_linear_regression.rb

@@ -26,7 +26,7 @@ class RubyLinearRegression
         @x = Matrix.rows( x_data )
         @y = Matrix.rows( y_data.collect { |e| [e] } )
 
-        @theta = Matrix[[0],[0]]
+        @theta = Matrix.zero(@x.column_count, 1)
   end
 
   # Compute the mean squared cost / error function
@@ -52,6 +52,26 @@ class RubyLinearRegression
     return @theta
   end
 
+  # Calculate optimal theta using gradient descent
+  # Arguments:
+  #   alpha: Learning rate
+  #   iterations: Number of iterations to run gradient descent
+  #   verbose: If true will output cost after each iteration, can be used to find optimal learning rate (alpha) and iteration
+  def train_gradient_descent( alpha = 0.01, iterations = 500, verbose = false )
+
+    0.upto( iterations ) do |i|
+      @temp_theta = Array.new(@theta.row_size)
+      0.upto(@theta.row_size-1) do |row|
+        @temp_theta[row] = @theta[row,0] - alpha * compute_gradient(row)
+      end
+
+      @theta = Matrix.columns([@temp_theta])
+
+      puts "Cost after #{i} iterations = #{compute_cost}" if verbose
+    end
+
+  end
+
   # Makes a prediction based on your trained model.
   # train_normal_equation must be called prior to making a prediction.
   #
@@ -102,4 +122,15 @@ class RubyLinearRegression
 
     end
 
+    # Compute the mean squared cost / error function
+    def compute_gradient( parameter )
+
+      # First use matrix multiplication and vector subtracton to find errors
+      gradients = ((@x * @theta) - @y).transpose * @x.column(parameter)
+
+      # Mean the grandient
+      mean = gradients.inject{ |sum, e| sum + e } / gradients.size
+
+      return mean
+    end
 end

metadata

@@ -1,19 +1,20 @@
 --- !ruby/object:Gem::Specification
 name: ruby_linear_regression
 version: !ruby/object:Gem::Version
-  version: 0.0.2
+  version: 0.1.0
 platform: ruby
 authors:
 - Soren Blond Daugaard
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2017-06-10 00:00:00.000000000 Z
+date: 2017-06-13 00:00:00.000000000 Z
 dependencies: []
 description: "An implementation of a linear regression machine learning algorithm
-  implemented in Ruby.\n                This algorithm uses Ruby's Matrix implementation
-  and the normal equation to train the data to the best fit.\n                The
-  algorithm works with multiple independent variables to predict a dependent variable. "
+  in Ruby.\n                This algorithm uses Ruby's Matrix implementation and the
+  normal equation to train the data to the best fit.\n                The algorithm
+  works with one independent variable or as a multivariate linear regression with
+  multiple variables to predict a dependent variable. "
 email: sbd@ineptum.dk
 executables: []
 extensions: []
@@ -40,7 +41,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.6.8
+rubygems_version: 2.6.12
 signing_key: 
 specification_version: 4
 summary: Linear regression implemented in Ruby.