add option for selecting gradient descent method for fitting linear regression

Author: kortirso

lib/learn_kit/regression/linear.ex

@@ -50,11 +50,16 @@ defmodule LearnKit.Regression.Linear do
   end
 
   @doc """
-  Fit train data
+  Fit train data with least squares method
 
   ## Parameters
 
     - predictor: %LearnKit.Regression.Linear{}
+    - options: keyword list with options
+
+  ## Options
+
+    - method: method for fit, "least squares"/"gradient descent", default is "least squares", optional
 
   ## Examples
 
@@ -65,11 +70,28 @@ defmodule LearnKit.Regression.Linear do
         results: [3, 6, 10, 15]
       }
 
+      iex> predictor = predictor |> LearnKit.Regression.Linear.fit([method: "gradient descent"])
+      %LearnKit.Regression.Linear{
+        coefficients: [-1.5, 4.0],
+        factors: [1, 2, 3, 4],
+        results: [3, 6, 10, 15]
+      }
+
   """
   @spec fit(%Linear{factors: factors, results: results}) :: %Linear{factors: factors, results: results, coefficients: coefficients}
 
-  def fit(%Linear{factors: factors, results: results}) do
-    %Linear{factors: factors, results: results, coefficients: fit_data(factors, results)}
+  def fit(%Linear{factors: factors, results: results}, options \\ []) do
+    coefficients = Keyword.merge([method: ""], options)
+                    |> define_method_for_fit
+                    |> fit_data(factors, results)
+    %Linear{factors: factors, results: results, coefficients: coefficients}
+  end
+
+  defp define_method_for_fit(options) do
+    cond do
+      Keyword.get(options, :method) == "gradient descent" -> "gradient descent"
+      true -> ""
+    end
   end
 
   @doc """

lib/learn_kit/regression/linear/fit.ex

@@ -7,7 +7,11 @@ defmodule LearnKit.Regression.Linear.Fit do
 
   defmacro __using__(_opts) do
     quote do
-      defp fit_data(factors, results) do
+      defp fit_data(method, factors, results) when method == "gradient descent" do
+
+      end
+
+      defp fit_data(_, factors, results) do
         beta = Math.correlation(factors, results) * Math.standard_deviation(results) / Math.standard_deviation(factors)
         alpha = Math.mean(results) - beta * Math.mean(factors)
         [alpha, beta]