eps 0.2.0 → 0.2.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 5ce1ae30056e3aaa08465d63571685ed2d43a17b40dc95bb2020d8d1f6752d6d
-  data.tar.gz: f6f9652d164991b82fed1375b9039baa6cbe3dd3b5413cdee244ba5aac923012
+  metadata.gz: ae5bc00818b79dc5e07f4dcda7ca56aa825f1014d1f70203564a87cb49b375d4
+  data.tar.gz: 8ba22dddc8635da418c429c12066c63bc1aea15238c32c4a7c4185f66281b6a5
 SHA512:
-  metadata.gz: bd9ddb1589b2866c42fda230389a27124285bd162ca9863528b41531d1c725fca3ac7327786b373e9cecda014d9cc2f34f791f203da77215f738cf92c261db61
-  data.tar.gz: 7d209acc346b223d00827c1df817a2017882b9446ff523f985e4c89b44f10f5a702d6c6dd957c8fb9ceccd83b98f7492aa15e029c1d0655787c1823348beacd0
+  metadata.gz: e8a0f8cc325d26618691613a6213f6471b45c94a22bb2c9eb6ea729543dce4deabd9875d8e7055649fde90066d30d09c0b1b61949598c6859557e8270ff8e776
+  data.tar.gz: 0b8d0918e9571ce1587d4497b8de84b64222f19ba7c466d24bd490605dcfbed10ebfa10ad10ed6f03a13ab45a7bc1b77d5928f45d0a9f8ba7757e569591a36fe

data/CHANGELOG.md

@@ -1,3 +1,9 @@
+## 0.2.1
+
+- Fixed error with `summary`
+- Fixed error with `predict` in `Eps::Base`
+- Fixed error with loaded classification models
+
 ## 0.2.0
 
 - Added support for classification

data/README.md

@@ -233,16 +233,17 @@ class PriceModel < Eps::Base
     metrics = model.evaluate(test_features, test_target)
     puts "Test RMSE: #{metrics[:rmse]}"
     # for classification, use:
-    # puts "Test accuracy: #{metrics[:accuracy]}"
+    # puts "Test accuracy: #{(100 * metrics[:accuracy]).round}%"
 
     # finalize
     houses = preprocess(houses)
     all_features = houses.map { |h| features(h) }
     all_target = houses.map { |h| target(h) }
-    @model = Eps::Model.new(all_features, all_target)
+    model = Eps::Model.new(all_features, all_target)
 
     # save
-    File.write(model_file, @model.to_pmml)
+    File.write(model_file, model.to_pmml)
+    @model = nil # reset for future predictions
   end
 
   def predict(house)

data/lib/eps/base.rb

@@ -1,12 +1,12 @@
 module Eps
   class Base
     class << self
-      def build
-        instance.build
+      def build(*args)
+        instance.build(*args)
       end
 
-      def predict
-        instance.predict
+      def predict(*args)
+        instance.predict(*args)
       end
 
       private

data/lib/eps/base_estimator.rb

@@ -12,15 +12,13 @@ module Eps
 
       y = prep_y(y.to_a)
 
-      @target = target || "target"
-
       if x.size != y.size
         raise "Number of samples differs from target"
       end
 
       @x = x
       @y = y
-      @target = target
+      @target = target || "target"
     end
 
     def predict(x)

data/lib/eps/linear_regression.rb

@@ -223,26 +223,25 @@ module Eps
 
     # https://people.richland.edu/james/ictcm/2004/multiple.html
     def summary(extended: false)
-      @summary_str ||= begin
-        str = String.new("")
-        len = [coefficients.keys.map(&:size).max, 15].max
+      coefficients = @coefficients
+      str = String.new("")
+      len = [coefficients.keys.map(&:size).max, 15].max
+      if extended
+        str += "%-#{len}s %12s %12s %12s %12s\n" % ["", "coef", "stderr", "t", "p"]
+      else
+        str += "%-#{len}s %12s %12s\n" % ["", "coef", "p"]
+      end
+      coefficients.each do |k, v|
         if extended
-          str += "%-#{len}s %12s %12s %12s %12s\n" % ["", "coef", "stderr", "t", "p"]
+          str += "%-#{len}s %12.2f %12.2f %12.2f %12.3f\n" % [display_field(k), v, std_err[k], t_value[k], p_value[k]]
         else
-          str += "%-#{len}s %12s %12s\n" % ["", "coef", "p"]
-        end
-        coefficients.each do |k, v|
-          if extended
-            str += "%-#{len}s %12.2f %12.2f %12.2f %12.3f\n" % [display_field(k), v, std_err[k], t_value[k], p_value[k]]
-          else
-            str += "%-#{len}s %12.2f %12.3f\n" % [display_field(k), v, p_value[k]]
-          end
+          str += "%-#{len}s %12.2f %12.3f\n" % [display_field(k), v, p_value[k]]
         end
-        str += "\n"
-        str += "r2: %.3f\n" % [r2] if extended
-        str += "adjusted r2: %.3f\n" % [adjusted_r2]
-        str
       end
+      str += "\n"
+      str += "r2: %.3f\n" % [r2] if extended
+      str += "adjusted r2: %.3f\n" % [adjusted_r2]
+      str
     end
 
     def r2
@@ -280,12 +279,12 @@ module Eps
     # add epsilon for perfect fits
     # consistent with GSL
     def t_value
-      @t_value ||= Hash[coefficients.map { |k, v| [k, v / (std_err[k] + Float::EPSILON)] }]
+      @t_value ||= Hash[@coefficients.map { |k, v| [k, v / (std_err[k] + Float::EPSILON)] }]
     end
 
     def p_value
       @p_value ||= begin
-        Hash[coefficients.map do |k, _|
+        Hash[@coefficients.map do |k, _|
           tp =
             if @gsl
               GSL::Cdf.tdist_P(t_value[k].abs, degrees_of_freedom)
@@ -350,7 +349,7 @@ module Eps
     end
 
     def degrees_of_freedom
-      @y.size - coefficients.size
+      @y.size - @coefficients.size
     end
 
     def mean(arr)

data/lib/eps/naive_bayes.rb

@@ -3,8 +3,8 @@ module Eps
     attr_reader :probabilities
 
     def initialize(probabilities: nil, target: nil)
-      @probabilities = probabilities if probabilities
-      @target = target if target
+      @probabilities = probabilities
+      @target = target
     end
 
     def train(*args)
@@ -22,7 +22,7 @@ module Eps
           xi[@target] = @y[i]
         end
         keys.each do |k|
-          conditional[k] = {}
+          conditional[k.to_s] = {}
           x.group_by { |xi| xi[@target] }.each do |group, xs|
             v = xs.map { |xi| xi[k] }
 
@@ -32,9 +32,9 @@ module Eps
               # 1. categorical features
               # 2. conditional probabilities
               # TODO more efficient count
-              conditional[k][group] = group_count(v)
+              conditional[k.to_s][group] = group_count(v)
             else
-              conditional[k][group] = {mean: mean(v), stdev: stdev(v)}
+              conditional[k.to_s][group] = {mean: mean(v), stdev: stdev(v)}
             end
           end
         end
@@ -48,15 +48,13 @@ module Eps
 
     # TODO better summary
     def summary(extended: false)
-      @summary_str ||= begin
-        str = String.new("")
-        probabilities[:prior].each do |k, v|
-          str += "#{k}: #{v}\n"
-        end
-        str += "\n"
-        str += "accuracy: %d%%\n" % [(100 * accuracy).round]
-        str
+      str = String.new("")
+      probabilities[:prior].each do |k, v|
+        str += "#{k}: #{v}\n"
       end
+      str += "\n"
+      str += "accuracy: %d%%\n" % [(100 * accuracy).round]
+      str
     end
 
     def accuracy
@@ -186,7 +184,7 @@ module Eps
 
     def _predict(x)
       x.map do |xi|
-        probs = calculate_class_probabilities(xi)
+        probs = calculate_class_probabilities(stringify_keys(xi))
         # deterministic for equal probabilities
         probs.sort_by { |k, v| [-v, k.to_s] }[0][0]
       end
@@ -236,5 +234,13 @@ module Eps
       sum = arr.inject(0) { |accum, i| accum + (i - m)**2 }
       Math.sqrt(sum / (arr.length - 1).to_f)
     end
+
+    def stringify_keys(h)
+      o = {}
+      h.each do |k, v|
+        o[k.to_s] = v
+      end
+      o
+    end
   end
 end

data/lib/eps/version.rb

@@ -1,3 +1,3 @@
 module Eps
-  VERSION = "0.2.0"
+  VERSION = "0.2.1"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: eps
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.2.1
 platform: ruby
 authors:
 - Andrew Kane