code refactoring, release 0.1.6

CHANGELOG.md

@@ -4,7 +4,7 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.html).
 
-## Unreleased
+## [0.1.6] - 2019-01-08
 ### Added
 - Polynomial Regression predictor
 

README.md

@@ -9,6 +9,7 @@ Available preprocessing methods:
 Available algorithms for prediction:
 
 - Linear Regression
+- Polynomial Regression
 
 Available algorithms for classification:
 
@@ -23,7 +24,7 @@ by adding `learn_kit` to your list of dependencies in `mix.exs`:
 ```elixir
 def deps do
   [
-    {:learn_kit, "~> 0.1.5"}
+    {:learn_kit, "~> 0.1.6"}
   ]
 end
 ```

lib/learn_kit/knn.ex

@@ -6,7 +6,6 @@ defmodule LearnKit.Knn do
   defstruct data_set: []
 
   alias LearnKit.Knn
-
   use Knn.Classify
 
   @type label :: atom

lib/learn_kit/knn/classify.ex

@@ -45,14 +45,10 @@ defmodule LearnKit.Knn.Classify do
       end
 
       # sort distances
-      defp sort_distances(features) do
-        Enum.sort(features, &(elem(&1, 0) <= elem(&2, 0)))
-      end
+      defp sort_distances(features), do: Enum.sort(features, &(elem(&1, 0) <= elem(&2, 0)))
 
       # take closest features
-      defp select_closest_features(features, options) do
-        Enum.take(features, options[:k])
-      end
+      defp select_closest_features(features, options), do: Enum.take(features, options[:k])
 
       # check existeness of current feature in data set
       defp check_zero_distance(closest_features, options) do
@@ -146,9 +142,7 @@ defmodule LearnKit.Knn.Classify do
         end
       end
 
-      defp accumulate_weight_of_labels([], acc) do
-        acc
-      end
+      defp accumulate_weight_of_labels([], acc), do: acc
 
       defp accumulate_weight_of_labels([{_, key, weight} | tail], acc) do
         previous = if Keyword.has_key?(acc, key), do: acc[key], else: 0

lib/learn_kit/naive_bayes/gaussian.ex

@@ -6,7 +6,6 @@ defmodule LearnKit.NaiveBayes.Gaussian do
   defstruct data_set: [], fit_data: []
 
   alias LearnKit.NaiveBayes.Gaussian
-
   use Gaussian.Normalize
   use Gaussian.Fit
   use Gaussian.Classify
@@ -34,9 +33,7 @@ defmodule LearnKit.NaiveBayes.Gaussian do
   """
   @spec new() :: %Gaussian{data_set: []}
 
-  def new do
-    Gaussian.new([])
-  end
+  def new, do: Gaussian.new([])
 
   @doc """
   Creates classifier with data_set
@@ -53,9 +50,7 @@ defmodule LearnKit.NaiveBayes.Gaussian do
   """
   @spec new(data_set) :: %Gaussian{data_set: data_set}
 
-  def new(data_set) do
-    %Gaussian{data_set: data_set}
-  end
+  def new(data_set), do: %Gaussian{data_set: data_set}
 
   @doc """
   Add train data to classifier

lib/learn_kit/preprocessing.ex

@@ -4,7 +4,6 @@ defmodule LearnKit.Preprocessing do
   """
 
   alias LearnKit.{Preprocessing, Math}
-
   use Preprocessing.Normalize
 
   @type row :: [number]

lib/learn_kit/regression/linear.ex

@@ -6,7 +6,6 @@ defmodule LearnKit.Regression.Linear do
   defstruct factors: [], results: [], coefficients: []
 
   alias LearnKit.Regression.Linear
-
   use Linear.Calculations
   use LearnKit.Regression.Score
 
@@ -43,8 +42,9 @@ defmodule LearnKit.Regression.Linear do
   """
   @spec new(factors, results) :: %Linear{factors: factors, results: results, coefficients: []}
 
-  def new(factors, results) when is_list(factors) and is_list(results),
-    do: %Linear{factors: factors, results: results}
+  def new(factors, results) when is_list(factors) and is_list(results) do
+    %Linear{factors: factors, results: results}
+  end
 
   @doc """
   Fit train data
@@ -75,19 +75,13 @@ defmodule LearnKit.Regression.Linear do
       }
 
   """
-  @spec fit(%Linear{factors: factors, results: results}) :: %Linear{
-          factors: factors,
-          results: results,
-          coefficients: coefficients
-        }
-
-  def fit(linear = %Linear{factors: factors, results: results}, options \\ [])
-      when is_list(options) do
+  @spec fit(%Linear{factors: factors, results: results}) :: %Linear{factors: factors, results: results, coefficients: coefficients}
+
+  def fit(linear = %Linear{factors: factors, results: results}, options \\ []) when is_list(options) do
     coefficients =
       Keyword.merge([method: ""], options)
       |> define_method_for_fit()
       |> do_fit(linear)
-
     %Linear{factors: factors, results: results, coefficients: coefficients}
   end
 

lib/learn_kit/regression/linear/calculations.ex

@@ -3,13 +3,11 @@ defmodule LearnKit.Regression.Linear.Calculations do
   Module for fit functions
   """
 
-  alias LearnKit.Math
-  alias LearnKit.Regression.Linear
+  alias LearnKit.{Math, Regression.Linear}
 
   defmacro __using__(_opts) do
     quote do
-      defp do_fit(method, %Linear{factors: factors, results: results})
-           when method == "gradient descent" do
+      defp do_fit("gradient descent", %Linear{factors: factors, results: results}) do
         gradient_descent_iteration(
           [:rand.uniform(), :rand.uniform()],
           0.0001,
@@ -21,10 +19,7 @@ defmodule LearnKit.Regression.Linear.Calculations do
       end
 
       defp do_fit(_, %Linear{factors: factors, results: results}) do
-        beta =
-          Math.correlation(factors, results) * Math.standard_deviation(results) /
-            Math.standard_deviation(factors)
-
+        beta = calc_beta(factors, results)
         alpha = Math.mean(results) - beta * Math.mean(factors)
         [alpha, beta]
       end
@@ -36,66 +31,54 @@ defmodule LearnKit.Regression.Linear.Calculations do
         end)
       end
 
+      defp calc_beta(factors, results) do
+        Math.correlation(factors, results) * Math.standard_deviation(results) / Math.standard_deviation(factors)
+      end
+
       defp squared_error_gradient(linear, x, y) do
         error_variable = prediction_error(linear, x, y)
-
         [
           -2 * error_variable,
           -2 * error_variable * x
         ]
       end
 
-      defp gradient_descent_iteration(_, _, min_theta, _, _, iterations_with_no_improvement)
-           when iterations_with_no_improvement >= 100,
-           do: min_theta
+      defp gradient_descent_iteration(_, _, min_theta, _, _, no_improve_step) when no_improve_step >= 100, do: min_theta
 
-      defp gradient_descent_iteration(
-             theta,
-             alpha,
-             min_theta,
-             min_value,
-             data,
-             iterations_with_no_improvement
-           ) do
+      defp gradient_descent_iteration(theta, alpha, min_theta, min_value, data, no_improve_step) do
         [
           min_theta,
           min_value,
-          iterations_with_no_improvement,
+          no_improve_step,
           alpha
-        ] = check_value(data, min_value, theta, min_theta, iterations_with_no_improvement, alpha)
+        ] = check_value(data, min_value, theta, min_theta, no_improve_step, alpha)
 
-        theta =
-          data
-          |> Enum.shuffle()
-          |> Enum.reduce(theta, fn {xi, yi}, acc ->
-            gradient_i = squared_error_gradient(%Linear{coefficients: theta}, xi, yi)
-            acc |> Math.vector_subtraction(alpha |> Math.scalar_multiply(gradient_i))
-          end)
-
-        gradient_descent_iteration(
-          theta,
-          alpha,
-          min_theta,
-          min_value,
-          data,
-          iterations_with_no_improvement
-        )
+        calc_new_theta(data, theta, alpha)
+        |> gradient_descent_iteration(alpha, min_theta, min_value, data, no_improve_step)
       end
 
-      defp check_value(data, min_value, theta, min_theta, iterations_with_no_improvement, alpha) do
-        value =
-          Enum.reduce(data, 0, fn {xi, yi}, acc ->
-            acc + squared_prediction_error(%Linear{coefficients: theta}, xi, yi)
-          end)
+      defp calc_new_theta(data, theta, alpha) do
+        data
+        |> Enum.shuffle()
+        |> Enum.reduce(theta, fn {xi, yi}, acc ->
+          gradient_i = squared_error_gradient(%Linear{coefficients: theta}, xi, yi)
+          acc |> Math.vector_subtraction(alpha |> Math.scalar_multiply(gradient_i))
+        end)
+      end
 
+      defp check_value(data, min_value, theta, min_theta, no_improve_step, alpha) do
+        value = calc_new_value(data, theta)
         cond do
-          value < min_value ->
-            [theta, value, 0, 0.0001]
-
-          true ->
-            [min_theta, min_value, iterations_with_no_improvement + 1, alpha * 0.9]
+          value < min_value -> [theta, value, 0, 0.0001]
+          true -> [min_theta, min_value, no_improve_step + 1, alpha * 0.9]
         end
       end
+
+      defp calc_new_value(data, theta) do
+        Enum.reduce(data, 0, fn {xi, yi}, acc ->
+          acc + squared_prediction_error(%Linear{coefficients: theta}, xi, yi)
+        end)
+      end
     end
   end
 end

lib/learn_kit/regression/polynomial.ex

@@ -6,7 +6,6 @@ defmodule LearnKit.Regression.Polynomial do
   defstruct factors: [], results: [], coefficients: [], degree: 2
 
   alias LearnKit.Regression.Polynomial
-
   use Polynomial.Calculations
   use LearnKit.Regression.Score
 
@@ -29,12 +28,8 @@ defmodule LearnKit.Regression.Polynomial do
       %LearnKit.Regression.Polynomial{factors: [1, 2, 3, 4], results: [3, 6, 10, 15], coefficients: [], degree: 2}
 
   """
-  @spec new(factors, results) :: %Polynomial{
-          factors: factors,
-          results: results,
-          coefficients: [],
-          degree: 2
-        }
+  @spec new(factors, results) :: %Polynomial{factors: factors, results: results, coefficients: [], degree: 2}
+
   def new(factors, results) when is_list(factors) and is_list(results) do
     %Polynomial{factors: factors, results: results}
   end
@@ -74,12 +69,8 @@ defmodule LearnKit.Regression.Polynomial do
       }
 
   """
-  @spec fit(%Polynomial{factors: factors, results: results}) :: %Polynomial{
-          factors: factors,
-          results: results,
-          coefficients: coefficients,
-          degree: degree
-        }
+  @spec fit(%Polynomial{factors: factors, results: results}) :: %Polynomial{factors: factors, results: results, coefficients: coefficients, degree: degree}
+
   def fit(%Polynomial{factors: factors, results: results}, options \\ []) do
     degree = options[:degree] || 2
     matrix = matrix(factors, degree)
@@ -103,8 +94,8 @@ defmodule LearnKit.Regression.Polynomial do
 
   """
   @spec predict(%Polynomial{coefficients: coefficients, degree: degree}, list) :: {:ok, list}
-  def predict(polynomial = %Polynomial{coefficients: _, degree: _}, samples)
-      when is_list(samples) do
+
+  def predict(polynomial = %Polynomial{coefficients: _, degree: _}, samples) when is_list(samples) do
     {:ok, do_predict(polynomial, samples)}
   end
 
@@ -123,6 +114,7 @@ defmodule LearnKit.Regression.Polynomial do
 
   """
   @spec predict(%Polynomial{coefficients: coefficients, degree: degree}, number) :: {:ok, number}
+
   def predict(%Polynomial{coefficients: coefficients, degree: degree}, sample) do
     ordered_coefficients = coefficients |> Enum.reverse()
     {:ok, substitute_coefficients(ordered_coefficients, sample, degree, 0.0)}

lib/learn_kit/regression/polynomial/calculations.ex

@@ -14,7 +14,6 @@ defmodule LearnKit.Regression.Polynomial.Calculations do
 
       defp matrix_line(1, factors, degree) do
         power_ofs = Enum.to_list(1..degree)
-
         [Enum.count(factors) | sum_of_x_i_with_k(power_ofs, factors)]
       end
 
@@ -26,7 +25,6 @@ defmodule LearnKit.Regression.Polynomial.Calculations do
 
       defp matrix(factors, degree) do
         lines = Enum.to_list(1..(degree + 1))
-
         Enum.map(lines, fn line ->
           matrix_line(line, factors, degree)
         end)

lib/learn_kit/regression/score.ex

@@ -20,11 +20,7 @@ defmodule LearnKit.Regression.Score do
           {:ok, 0.9876543209876543}
 
       """
-      @spec score(%LearnKit.Regression.Linear{
-              factors: factors,
-              results: results,
-              coefficients: coefficients
-            }) :: {:ok, number}
+      @spec score(%LearnKit.Regression.Linear{factors: factors, results: results, coefficients: coefficients}) :: {:ok, number}
 
       def score(regression = %_{factors: _, results: _, coefficients: _}) do
         {
@@ -44,10 +40,9 @@ defmodule LearnKit.Regression.Score do
         y - prediction
       end
 
-      defp sum_of_squared_errors(
-             regression = %_{coefficients: _, factors: factors, results: results}
-           ) do
-        Enum.zip(factors, results)
+      defp sum_of_squared_errors(regression = %_{coefficients: _, factors: factors, results: results}) do
+        factors
+        |> Enum.zip(results)
         |> Enum.reduce(0, fn {xi, yi}, acc ->
           acc + squared_prediction_error(regression, xi, yi)
         end)

mix.exs

@@ -8,7 +8,7 @@ defmodule LearnKit.MixProject do
   def project do
     [
       app: :learn_kit,
-      version: "0.1.5",
+      version: "0.1.6",
       elixir: "~> 1.7",
       name: "LearnKit",
       description: @description,

test/learn_kit/knn_test.exs

@@ -1,6 +1,5 @@
 defmodule LearnKit.KnnTest do
   use ExUnit.Case
-
   alias LearnKit.Knn
 
   setup_all do

test/learn_kit/math_test.exs

@@ -1,6 +1,5 @@
 defmodule LearnKit.MathTest do
   use ExUnit.Case
-
   alias LearnKit.Math
 
   test "calculate sum" do

test/learn_kit/preprocessing_test.exs

@@ -1,6 +1,5 @@
 defmodule LearnKit.PreprocessingTest do
   use ExUnit.Case
-
   alias LearnKit.Preprocessing
 
   describe "for invalid data" do