daru 0.0.4 → 0.0.5

data/spec/fixtures/sales-funnel.csv

@@ -0,0 +1,18 @@
+Account,Name,Rep,Manager,Product,Quantity,Price,Status
+714466,Trantow-Barrows,Craig Booker,Debra Henley,CPU,1,30000,presented
+714466,Trantow-Barrows,Craig Booker,Debra Henley,Software,1,10000,presented
+714466,Trantow-Barrows,Craig Booker,Debra Henley,Maintenance,2,5000,pending
+737550,"Fritsch, Russel and Anderson",Craig Booker,Debra Henley,CPU,1,35000,declined
+146832,Kiehn-Spinka,Daniel Hilton,Debra Henley,CPU,2,65000,won
+218895,Kulas Inc,Daniel Hilton,Debra Henley,CPU,2,40000,pending
+218895,Kulas Inc,Daniel Hilton,Debra Henley,Software,1,10000,presented
+412290,Jerde-Hilpert,John Smith,Debra Henley,Maintenance,2,5000,pending
+740150,Barton LLC,John Smith,Debra Henley,CPU,1,35000,declined
+141962,Herman LLC,Cedric Moss,Fred Anderson,CPU,2,65000,won
+163416,Purdy-Kunde,Cedric Moss,Fred Anderson,CPU,1,30000,presented
+239344,Stokes LLC,Cedric Moss,Fred Anderson,Maintenance,1,5000,pending
+239344,Stokes LLC,Cedric Moss,Fred Anderson,Software,1,10000,presented
+307599,"Kassulke, Ondricka and Metz",Wendy Yule,Fred Anderson,Maintenance,3,7000,won
+688981,Keeling LLC,Wendy Yule,Fred Anderson,CPU,5,100000,won
+729833,Koepp Ltd,Wendy Yule,Fred Anderson,CPU,2,65000,declined
+729833,Koepp Ltd,Wendy Yule,Fred Anderson,Monitor,2,5000,presented

data/spec/index_spec.rb

@@ -17,25 +17,6 @@ describe Daru::Index do
     end
   end
 
-  context "#re_index" do
-    before :each do
-      @old = Daru::Index.new [:bob, :fisher, :zakir]
-    end
-    it "returns a new index object" do
-      n   = @old.re_index(@old + [:john, :shrinivas]) 
-
-      expect(n.object_id).not_to eq(@old.object_id)
-      expect(n.to_a).to eq([:bob, :fisher, :zakir, :john, :shrinivas])
-    end
-
-    it "does not over-ride existing indexes" do
-      n = @old.re_index(@old + :bob)
-
-      expect(n.object_id).not_to eq(@old.object_id)
-      expect(n.to_a)     .to eq([:bob, :fisher, :zakir])
-    end
-  end
-
   context "#+" do
     before :each do
       @left = Daru::Index.new [:miles, :geddy, :eric]
@@ -53,10 +34,18 @@ describe Daru::Index do
   end
 
   context "#[]" do
+    before do
+      @id = Daru::Index.new [:one, :two, :three, :four, :five, :six, :seven]
+    end
+
     it "works with ranges" do
-      id = Daru::Index.new [:one, :two, :three, :four, :five, :six, :seven]
+      expect(@id[:two..:five]).to eq(Daru::Index.new([:two, :three, :four, :five], 
+        [1,2,3,4]))
+    end
 
-      expect(id[:two..:five]).to eq(Daru::Index.new([:two, :three, :four, :five]))
+    it "returns multiple keys if specified multiple indices" do
+      expect(@id[[0,1,3,4]]).to eq(Daru::Index.new([:one, :two, :four, :five], 
+        [0,1,3,4]))
     end
   end
 end

data/spec/io/io_spec.rb

@@ -4,21 +4,11 @@ describe Daru::DataFrame do
   context ".from_csv" do
     it "loads from a CSV file" do
       df = Daru::DataFrame.from_csv('spec/fixtures/matrix_test.csv', 
-        col_sep: ' ', headers: true) do |csv|
-        csv.convert do |field, info|
-          case info[:header]
-          when :true_transform
-            field.split(',').map { |s| s.to_f }
-          else
-            field
-          end
-        end
-      end
-
-
-      expect(df.vectors).to eq([:image_resolution, :true_transform, :mls].to_index)
+        col_sep: ' ', headers: true)
+
+      expect(df.vectors).to eq([:image_resolution, :mls, :true_transform].to_index)
       expect(df.vector[:image_resolution].first).to eq(6.55779)
-      expect(df.vector[:true_transform].first[15]).to eq(1.0)
+      expect(df.vector[:true_transform].first).to eq("-0.2362347,0.6308649,0.7390552,0,0.6523478,-0.4607318,0.6018043,0,0.7201635,0.6242881,-0.3027024,4262.65,0,0,0,1")
     end
   end
 

data/spec/math/arithmetic/dataframe_spec.rb

@@ -1,5 +1,71 @@
 require 'spec_helper.rb'
 
 describe Daru::DataFrame do
+  before(:each) do
+    @df = Daru::DataFrame.new({a: [1,2,3,4,5], b: ['a','e','i','o','u'], 
+      c: [10,20,30,40,50]})
+    @left = Daru::DataFrame.new({a: [1,nil,nil,4], b: [10,nil,nil,40], c: [5,6,7,8]},
+      index: [0,4,5,3])
+    @right = Daru::DataFrame.new({a: [1,2,3,4,5], b: [10,20,30,40,50]}, 
+      index: [0,1,2,3,6])
+  end
 
+  context "#+" do
+    it "adds a number to all numeric vectors" do
+      expect(@df + 2).to eq(Daru::DataFrame.new({a: [3,4,5,6,7], b: ['a','e','i','o','u'], 
+      c: [12,22,32,42,52] }))
+    end
+
+    it "adds two dataframes to produce a third" do
+      expect(@left + @right).to eq(Daru::DataFrame.new({a: [2,nil,nil,8,nil,nil,nil],
+        b: [20,nil,nil,80,nil,nil,nil], c: [nil,nil,nil,nil,nil,nil]}, index: 
+        [0,1,2,3,4,5,6]))
+    end
+  end
+
+  context "#-" do
+    it "subtracts a number from all numeric vectors" do
+      expect(@df - 2).to eq(Daru::DataFrame.new({a: [-1,0,1,2,3], b: ['a','e','i','o','u'], 
+      c: [8,18,28,38,48]}))
+    end
+
+    it "subtracts a data frame from another" do
+
+    end
+  end
+
+  context "#*" do
+    it "multiplies a number with a DataFrame" do
+    end
+  end
+
+  context "#/" do
+
+  end
+
+  context "#%" do
+
+  end
+
+  context "#**" do
+
+  end
+
+  context "#sqrt" do
+    it "calculates sqrt" do
+      @df.sqrt
+    end
+  end
+
+  context "#round" do
+    it "rounds to precision" do
+      @df.round
+    end
+  end
+
+  context "#exp" do
+    it "calculates exponential" do
+      @df.exp
+    end
+  end
 end

data/spec/math/arithmetic/vector_spec.rb

@@ -4,21 +4,31 @@ describe Daru::Vector do
   before :each do
     @dv1 = Daru::Vector.new [1,2,3,4], name: :boozy, index: [:bud, :kf, :henie, :corona]
     @dv2 = Daru::Vector.new [1,2,3,4], name: :mayer, index: [:obi, :wan, :kf, :corona]
+    @with_md1 = Daru::Vector.new [1,2,3,nil,5,nil], name: :missing, index: [:a, :b, :c, :obi, :wan, :corona]
+    @with_md2 = Daru::Vector.new [1,2,3,nil,5,nil], name: :missing, index: [:obi, :wan, :corona, :a, :b, :c]
   end
 
   context "#+" do
     it "adds matching indexes of the other vector" do
-      expect(@dv1 + @dv2).to eq(Daru::Vector.new([5, 8], name: :boozy, index: [:kf, :corona]))
+      expect(@dv1 + @dv2).to eq(Daru::Vector.new([nil,8,nil,5,nil,nil], name: :boozy, index: [:bud,:corona,:henie,:kf,:obi,:wan]))
     end
 
     it "adds number to each element of the entire vector" do
       expect(@dv1 + 5).to eq(Daru::Vector.new [6,7,8,9], name: :boozy, index: [:bud, :kf, :henie, :corona])
     end
+
+    it "does not add when a number is being added" do
+      expect(@with_md1 + 1).to eq(Daru::Vector.new([2,3,4,nil,6,nil], name: :missing, index: [:a, :b, :c, :obi, :wan, :corona]))
+    end
+
+    it "puts a nil when one of the operands is nil" do
+      expect(@with_md1 + @with_md2).to eq(Daru::Vector.new([nil,7,nil,nil,nil,7], name: :missing, index: [:a, :b, :c, :corona, :obi, :wan]))
+    end
   end
 
   context "#-" do
     it "subtracts matching indexes of the other vector" do
-      expect(@dv1 - @dv2).to eq(Daru::Vector.new([-1,0], name: :boozy, index: [:kf, :corona]))
+      expect(@dv1 - @dv2).to eq(Daru::Vector.new([nil,0,nil,-1,nil,nil], name: :boozy, index: [:bud,:corona,:henie,:kf,:obi,:wan]))
     end
 
     it "subtracts number from each element of the entire vector" do
@@ -26,7 +36,7 @@ describe Daru::Vector do
     end
   end
 
-  context "#*"
+  context "#*" do
     it "multiplies matching indexes of the other vector" do
 
     end
@@ -48,4 +58,35 @@ describe Daru::Vector do
 
   context "#%" do
 
-  end
+  end
+
+  context "#**" do
+    
+  end
+
+  context "#exp" do
+    it "calculates exp of all numbers" do
+      expect(@with_md1.exp.round(3)).to eq(Daru::Vector.new([2.718281828459045, 
+        7.38905609893065, 20.085536923187668, nil, 148.4131591025766, nil], index: 
+        [:a, :b, :c, :obi, :wan, :corona], name: :missing).round(3))
+    end
+  end
+
+  context "#abs" do
+    it "calculates abs value" do
+      @with_md1.abs
+    end
+  end
+
+  context "#sqrt" do
+    it "calculates sqrt" do
+      @with_md1.sqrt
+    end
+  end
+
+  context "#round" do
+    it "rounds to given precision" do
+      @with_md1.round(2)
+    end 
+  end
+end

data/spec/math/statistics/dataframe_spec.rb

@@ -1,5 +1,95 @@
 require 'spec_helper.rb'
 
 describe Daru::DataFrame do
-  
+  before do
+    @df = Daru::DataFrame.new({
+      a: ['foo'  ,  'foo',  'foo',  'foo',  'foo',  'bar',  'bar',  'bar',  'bar'], 
+      b: ['one'  ,  'one',  'one',  'two',  'two',  'one',  'one',  'two',  'two'],
+      c: ['small','large','large','small','small','large','small','large','small'],
+      d: [1,2,2,3,3,4,5,6,7],
+      e: [2,4,4,6,6,8,10,12,14],
+      f: [10,20,20,30,30,40,50,60,70]
+    })
+  end
+
+  context "#mean" do
+    it "calculates mean of single level numeric only vectors and returns values in a Vector" do
+      expect(@df.mean.round(2)).to eq(Daru::Vector.new([3.67, 7.33, 36.67], 
+        index: [:d, :e, :f]
+      ))
+    end
+
+    it "calculates mean of multi level numeric only vectors and returns values in a DataFrame" do
+      # TODO - pending
+    end
+  end
+
+  context "#std" do
+    it "calculates standard deviation of single leavel numeric only vectors and returns values in a Vector" do
+      expect(@df.std).to eq(Daru::Vector.new([2, 4, 20], index: [:d, :e, :f]))
+    end
+  end
+
+  context "#sum" do
+    it "calculates sum of single level numeric only vectors and returns values in a Vector" do
+      # TODO - write tests
+    end
+  end
+
+  context "#count" do
+    # TODO
+  end
+
+  context "#mode" do
+    # TODO
+  end
+
+  context "#median" do
+    # TODO
+  end
+
+  context "#max" do
+    # TODO
+  end
+
+  context "#min" do
+    # TODO
+  end
+
+  context "#product" do
+    # TODO
+  end
+
+  context "#describe" do
+    it "generates mean, std, max, min and count of numeric vectors in one shot" do
+      expect(@df.describe.round(2)).to eq(Daru::DataFrame.new({
+        d: [9.00, 3.67 ,2.00 , 1.00,  7.00],
+        e: [9.00, 7.33 ,4.00 , 2.00, 14.00],
+        f: [9.00, 36.67,20.00,10.00, 70.00]
+        }, index: [:count, :mean, :std, :min, :max]
+      ))
+    end
+  end
+
+  context "#cov" do
+    it "calculates the variance covariance of the numeric vectors of DataFrame" do
+      expect(@df.cov).to eq(Daru::DataFrame.new({
+        d: [4,8,40],
+        e: [8,16,80],
+        f: [40,80,400]
+        }, index: [:d, :e, :f]
+      ))
+    end
+  end
+
+  context "#corr", focus: true do
+    it "calculates the correlation between the numeric vectors of DataFrame" do
+      expect(@df.corr).to eq(Daru::DataFrame.new({
+        d: [1,1,1],
+        e: [1,1,1],
+        f: [1,1,1]
+        }, index: [:d, :e, :f]
+      ))
+    end
+  end
 end

data/spec/math/statistics/vector_spec.rb

@@ -1,39 +1,41 @@
 require 'spec_helper.rb'
 
 describe Daru::Vector do
-  [NMatrix, Array].each do |dtype|
+  [:array, :nmatrix].each do |dtype|
     describe dtype do
       before :each do
         @dv = Daru::Vector.new [323, 11, 555, 666, 234, 21, 666, 343, 1, 2], dtype: dtype
+        @dv_with_md = Daru::Vector.new [323, 11, 555, nil, 666, 234, 21, 666, 343, nil, 1, 2]
       end
 
       context "#mean" do
         it "calculates mean" do
           expect(@dv.mean).to eq(282.2)
+          expect(@dv_with_md.mean).to eq(282.2)
         end
       end
 
       context "#sum_of_squares" do
         it "calcs sum of squares" do
-
+          @dv.sum_of_squares
         end
       end
 
       context "#standard_deviation_sample" do
         it "calcs standard deviation sample" do
-
+          @dv.standard_deviation_sample
         end
       end
 
       context "#variance_sample" do
         it "calculates sample variance" do
-
+          @dv.variance_sample
         end
       end
 
       context "#standard_deviation_population" do
         it "calculates standard deviation population" do
-
+          @dv.standard_deviation_population
         end
       end
 
@@ -124,7 +126,9 @@ describe Daru::Vector do
       end
 
       context "#proportion" do
-
+        it "calculates proportion" do
+          @dv.proportion
+        end
       end
 
       context "#proportions" do
@@ -140,7 +144,13 @@ describe Daru::Vector do
       end
 
       context "#count" do
+        it "counts specified element" do
+          @dv.count(323)
+        end
 
+        it "counts total number of elements" do
+          expect(@dv.count).to eq(10)
+        end
       end
 
       context "#coefficient_of_variation" do
@@ -148,6 +158,22 @@ describe Daru::Vector do
           @dv.coefficient_of_variation
         end
       end
+
+      context "#factor" do
+
+      end
+
+      context "#median_absolute_deviation" do
+        it "calculates median_absolute_deviation" do
+          @dv.median_absolute_deviation
+        end
+      end
+
+      context "#standard_error" do
+        it "calculates standard error" do
+          @dv.standard_error
+        end
+      end
     end
   end
 end

data/spec/monkeys_spec.rb

@@ -1,6 +1,15 @@
 require 'spec_helper.rb'
 
 describe "Monkeys" do
-  context "Array" do
+  context Array do
+  end
+
+  context Matrix do
+    it "performs elementwise division" do
+      left  = Matrix[[3,6,9],[4,8,12],[2,4,6]]
+      right = Matrix[[3,6,9],[4,8,12],[2,4,6]]
+
+      expect(left.elementwise_division(right)).to eq(Matrix[[1,1,1],[1,1,1],[1,1,1]])
+    end
   end
 end