daru 0.1.6 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 10338f8d554cc2c70b2dcc2d8fd029e73446f4de
-  data.tar.gz: b848b5923eebe90577ef93d4c9e988d7aa09fe9b
+  metadata.gz: 69452b32fd8ef0ef7fb4ed58ab53ffa8aa15806d
+  data.tar.gz: 56927c77adbe7941eb2ca9a5e44d705931aad237
 SHA512:
-  metadata.gz: ad6c0de4217a65a2c6245c4b969e98d970da47a08ce8128b060cd06a13ae415df6d5e7fdc7bba4c84115fc8559e80bea0a39c87efb1c3401df358f4df5d43117
-  data.tar.gz: 34abe4afd0c88c24d3d0bbd6f11df5efe554f50b0d3d045a8a969f4eae2232d8d1cfeead814e336da507ca94b3126447631699258fe1a5dc5ea736de77c587f8
+  metadata.gz: 8e7511133b3409f7821cfec944a950d53df57bcd5893bb8a9557c013f31bf1e4a9cc07bbe1c143c63684f00f7d8d8f1adf3b31df732508e667ba6677f47d1d96
+  data.tar.gz: fc4beb70106372a276b21e0da645951595e5674f56e4422752aeeabc9cc2156983add90e59486aea4d88386fbeb2896d15f7ede30667bc84027abd900ee42e0e

data/.github/ISSUE_TEMPLATE.md

@@ -0,0 +1,18 @@
+Heya! We are glad you are going to contribute to Daru by creating an issue, and kindly ask you to
+follow the simple rules:
+
+1. If it is a bug report, please provide a **self-containing** Ruby code for reproducing the bug.
+  This means if Daru contributors just copy-paste the code from issue into `this-is-bug.rb` and run
+  `ruby this-is-bug.rb`, it will be reproduced. If the bug is hard to spot (e.g. it is not some
+  `NoMethodError`, but the differences in data structure), please show it with comment in code or
+  plain text in the issue.
+2. If it is a feature request, try to do the following (if possible):
+  * show how new feature will work with small code example;
+  * explain the use case (if it is not 200% obvious);
+  * if you are aware of it, show how it works in pandas and/or R.
+3. If it is just a question ("how to do this or that" or "why Daru does this or that") feel free to
+  write it in any form that is convenient to you, but remember code examples and use cases are always
+  welcome.
+
+Thanks! And please remove this text when finished with your issue description :)
+

data/.rubocop.yml

@@ -12,6 +12,7 @@ AllCops:
     - 'vendor/**/*'
     - 'benchmarks/*'
     - 'profile/*'
+    - 'tmp/*'
   DisplayCopNames: true
   TargetRubyVersion: 2.0
 

data/.travis.yml

@@ -9,12 +9,17 @@ rvm:
   - '2.4.0'
 
 matrix:
+  allow_failures:
+    - rvm: '2.0'
   fast_finish:
     true
 
 script:
+  - bundle add yard-junk
+  - bundle install
   - bundle exec rspec
   - bundle exec rubocop
+  - bundle exec yard-junk
 
 install:
   - gem install bundler

data/History.md

@@ -1,3 +1,31 @@
+# 0.2.0 (31 October 2017)
+* Major Enhancements
+  - Add `DataFrame#which` query DSL (experimental! @rainchen)
+  - Add `DataFrame/Vector#rolling_fillna` (@baarkerlounger)
+  - Add `GroupBy#aggregate` (@shekharrajak)
+  - Add `DataFrame#uniq` (@baarkerlounger)
+
+* Minor Enhancements
+  - Allow `Vector#count` to be called without param for category type Vector (@rainchen)
+  - Add option to `DataFrame#vector_sum` to skip nils (@parthm)
+  - Add installation instructions to README.md (@koishimasato)
+  - Add release policy documentation (@baarkerlounger)
+  - Set index as DataFrame's default x axis for nyaplot (@matugm)
+
+* Fixes
+  - Fix `DataFrame/Vector#to_s` when name is a symbol (@baarkerlounger)
+  - Force `Vector#proportions` to return float (@rainchen)
+  - `DataFrame#new` creates empty DataFrame when given empty hash (@parthm)
+  - Remove unnecessary backports dependencies (@zverok)
+  - Specify minimum packable dependency (@zverok)
+  - Preserve key/column order when creating DataFrame from hash (@baarkerlounger)
+  - Fix `DataFrame#add_row` for DF with multi-index (@zverok)
+  - Fix `Vector#min, `#max`, `#index_of_min`, `#index_of_max` (0.1.6 regression) (@athityakumar)
+  - Integrate yard-junk into CI (@rohitner)
+  - Remove Travis spec restriction (@zverok)
+  - Fix tuple sorting for DataFrames with nils (@baarkerlounger)
+  - Fix merge on index dropping default index (@rohitner)
+
 # 0.1.6 (04 August 2017)
 * Major Enhancements
   - Add support for reading HTML tables into DataFrames (@athityakumar)

data/README.md

@@ -26,6 +26,12 @@ daru makes it easy and intuitive to process data predominantly through 2 data st
 * Quickly reducing data with pivot tables for quick data summary.
 * Import and export data from and to Excel, CSV, SQL Databases, ActiveRecord and plain text files.
 
+## Installation
+
+```console
+$ gem install daru
+```
+
 ## Notebooks
 
 #### Notebooks on most use cases

data/ReleasePolicy.md

@@ -0,0 +1,20 @@
+# Gem Release Policy
+
+Applicable to Daru > 0.1.6
+
+## Versioning
+
+Daru follows semantic versioning whereby the version number is always in the form MAJOR.MINOR.PATCH
+
+* Patch bump = Bug fixes
+* Minor bump = New features but backwards compatible
+* Major bump = API breaking changes
+
+For Major and Minor bumps release candidates should be released around 2 weeks prior to the bump and are indicated by MAJOR.MINOR.0.rc.
+
+For more information see the full semantic versioning specification at http://semver.org/.
+
+## Release Timing
+
+Patch releases should be done after every fix of a major bug (as tagged in the github issue tracker).
+Major releases should be kept to the minimum.

data/daru.gemspec

@@ -52,6 +52,9 @@ EOF
 
   spec.add_runtime_dependency 'backports'
 
+  # it is required by NMatrix, yet we want to specify clearly which minimal version is OK
+  spec.add_runtime_dependency 'packable', '~> 1.3.9'
+
   spec.add_development_dependency 'spreadsheet', '~> 1.1.1'
   spec.add_development_dependency 'bundler', '~> 1.10'
   spec.add_development_dependency 'rake', '~>10.5'
@@ -75,6 +78,7 @@ EOF
   spec.add_development_dependency 'simplecov'
   spec.add_development_dependency 'gruff'
   spec.add_development_dependency 'webmock'
+
   if RUBY_VERSION < '2.1.0'
     spec.add_development_dependency 'nokogiri', '<= 1.6.8.1'
   else

data/lib/daru.rb

@@ -105,6 +105,7 @@ require 'date'
 require 'daru/version.rb'
 
 require 'open-uri'
+require 'backports/2.1.0/array/to_h'
 
 require 'daru/index/index.rb'
 require 'daru/index/multi_index.rb'
@@ -124,5 +125,3 @@ require 'daru/core/merge.rb'
 
 require 'daru/date_time/offsets.rb'
 require 'daru/date_time/index.rb'
-
-require 'backports'

data/lib/daru/category.rb

@@ -1,5 +1,7 @@
 module Daru
   module Category # rubocop:disable Metrics/ModuleLength
+    UNDEFINED = Object.new.freeze
+
     attr_accessor :base_category
     attr_reader :index, :coding_scheme, :name
 
@@ -113,7 +115,7 @@ module Daru
     end
 
     # Associates a category to the vector.
-    # @param [Array] *new_categories new categories to be associated
+    # @param [Array] new_categories new categories to be associated
     # @example
     #   dv = Daru::Vector.new [:a, 1, :a, 1, :c], type: :category
     #   dv.add_category :b
@@ -131,7 +133,10 @@ module Daru
     #   dv = Daru::Vector.new [:a, 1, :a, 1, :c], type: :category
     #   dv.count :a
     #   # => 2
-    def count category
+    #   dv.count
+    #   # => 5
+    def count category=UNDEFINED
+      return @cat_hash.values.map(&:size).inject(&:+) if category == UNDEFINED # count all
       raise ArgumentError, "Invalid category #{category}" unless
         categories.include?(category)
 
@@ -167,7 +172,7 @@ module Daru
     end
 
     # Returns vector for indexes/positions specified
-    # @param [Array] *indexes indexes/positions for which values has to be retrived
+    # @param [Array] indexes for which values has to be retrived
     # @note Since it accepts both indexes and postions. In case of collision,
     #   arguement will be treated as index
     # @return vector containing values specified at specified indexes/positions
@@ -196,7 +201,7 @@ module Daru
     end
 
     # Returns vector for positions specified.
-    # @param [Array] *positions positions at which values to be retrived.
+    # @param [Array] positions at which values to be retrived.
     # @return vector containing values specified at specified positions
     # @example
     #   dv = Daru::Vector.new [:a, 1, :a, 1, :c], type: :category
@@ -223,7 +228,7 @@ module Daru
 
     # Modifies values at specified indexes/positions.
     # @note In order to add a new category you need to associate it via #add_category
-    # @param [Array] *indexes indexes/positions at which to modify value
+    # @param [Array] indexes at which to modify value
     # @param [object] val value to assign at specific indexes/positions
     # @return modified vector
     # @example
@@ -584,7 +589,7 @@ module Daru
     alias :gteq :mteq
 
     # For querying the data
-    # @param [object] arel like query syntax
+    # @param bool_array [object] arel like query syntax
     # @return [Daru::Vector] Vector which makes the conditions true
     # @example
     #   dv = Daru::Vector.new ['I', 'II', 'I', 'III', 'I', 'II'],
@@ -658,7 +663,7 @@ module Daru
     end
 
     # Check if any one of mentioned values occur in the vector
-    # @param [Array] *values values to check for
+    # @param [Array] values to check for
     # @return [true, false] returns true if any one of specified values
     #   occur in the vector
     # @example
@@ -670,7 +675,7 @@ module Daru
     end
 
     # Return a vector with specified values removed
-    # @param [Array] *values values to reject from resultant vector
+    # @param [Array] values to reject from resultant vector
     # @return [Daru::Vector] vector with specified values removed
     # @example
     #   dv = Daru::Vector.new [1, 2, nil, Float::NAN], type: :category
@@ -689,7 +694,7 @@ module Daru
     end
 
     # Count the number of values specified
-    # @param [Array] *values values to count for
+    # @param [Array] values to count for
     # @return [Integer] the number of times the values mentioned occurs
     # @example
     #   dv = Daru::Vector.new [1, 2, 1, 2, 3, 4, nil, nil]
@@ -702,7 +707,7 @@ module Daru
     end
 
     # Return indexes of values specified
-    # @param [Array] *values values to find indexes for
+    # @param [Array] values to find indexes for
     # @return [Array] array of indexes of values specified
     # @example
     #   dv = Daru::Vector.new [1, 2, nil, Float::NAN], index: 11..14

data/lib/daru/core/group_by.rb

@@ -11,12 +11,14 @@ module Daru
         end
       end
 
-      TUPLE_SORTER = lambda do |a, b|
-        if a && b
-          a.compact <=> b.compact
-        else
-          a ? 1 : -1
-        end
+      TUPLE_SORTER = lambda do |left, right|
+        return -1 unless right
+        return 1 unless left
+
+        left = left.compact
+        right = right.compact
+        return left <=> right || 0 if left.length == right.length
+        left.length <=> right.length
       end
 
       def initialize context, names
@@ -203,8 +205,8 @@ module Daru
 
       # Iteratively applies a function to the values in a group and accumulates the result.
       # @param init (nil) The initial value of the accumulator.
-      # @param block [Proc] A proc or lambda that accepts two arguments.  The first argument
-      #                     is the accumulated result.  The second argument is a DataFrame row.
+      # @yieldparam block [Proc] A proc or lambda that accepts two arguments.  The first argument
+      #                          is the accumulated result.  The second argument is a DataFrame row.
       # @example Usage of reduce
       #   df = Daru::DataFrame.new({
       #     a: ['a','b'] * 3,
@@ -243,6 +245,47 @@ module Daru
         @df.inspect
       end
 
+      # Function to use for aggregating the data.
+      # `group_by` is using Daru::DataFrame#aggregate
+      #
+      # @param options [Hash] options for column, you want in resultant dataframe
+      #
+      # @return [Daru::DataFrame]
+      #
+      # @example
+      #
+      #   df = Daru::DataFrame.new(
+      #     name: ['Ram','Krishna','Ram','Krishna','Krishna'],
+      #     visited: ['Hyderabad', 'Delhi', 'Mumbai', 'Raipur', 'Banglore'])
+      #
+      #   => #<Daru::DataFrame(5x2)>
+      #                   name   visited
+      #            0       Ram Hyderabad
+      #            1   Krishna     Delhi
+      #            2       Ram    Mumbai
+      #            3   Krishna    Raipur
+      #            4   Krishna  Banglore
+      #
+      #   df.group_by(:name)
+      #   => #<Daru::DataFrame(5x1)>
+      #                          visited
+      #      Krishna         1     Delhi
+      #                      3    Raipur
+      #                      4  Banglore
+      #          Ram         0 Hyderabad
+      #                      2    Mumbai
+      #
+      #   df.group_by(:name).aggregate(visited: -> (vec){vec.to_a.join(',')})
+      #   => #<Daru::DataFrame(2x1)>
+      #                  visited
+      #       Krishna Delhi,Raipur,Banglore
+      #           Ram Hyderabad,Mumbai
+      #
+      def aggregate(options={})
+        @df.index = @df.index.remove_layer(@df.index.levels.size - 1)
+        @df.aggregate(options)
+      end
+
       private
 
       def init_groups_df tuples, names

data/lib/daru/dataframe.rb

@@ -84,7 +84,7 @@ module Daru
       # Read a dataframe from AR::Relation
       #
       # @param relation [ActiveRecord::Relation] An AR::Relation object from which data is loaded
-      # @params fields [Array] Field names to be loaded (optional)
+      # @param fields [Array] Field names to be loaded (optional)
       #
       # @return A dataframe containing the data loaded from the relation
       #
@@ -277,6 +277,17 @@ module Daru
     # Default to *true*.
     #
     # == Usage
+    #
+    #   df = Daru::DataFrame.new
+    #   # =>
+    #   # <Daru::DataFrame(0x0)>
+    #   # Creates an empty DataFrame with no rows or columns.
+    #
+    #   df = Daru::DataFrame.new({}, order: [:a, :b])
+    #   #<Daru::DataFrame(0x2)>
+    #     a   b
+    #   # Creates a DataFrame with no rows and columns :a and :b
+    #
     #   df = Daru::DataFrame.new({a: [1,2,3,4], b: [6,7,8,9]}, order: [:b, :a],
     #     index: [:a, :b, :c, :d], name: :spider_man)
     #
@@ -329,7 +340,7 @@ module Daru
     #    #    1   4  14  44
     #    #    2   5  15  55
 
-    def initialize source, opts={} # rubocop:disable Metrics/MethodLength
+    def initialize source={}, opts={} # rubocop:disable Metrics/MethodLength
       vectors, index = opts[:order], opts[:index] # FIXME: just keyword arges after Ruby 2.1
       @data = []
       @name = opts[:name]
@@ -375,7 +386,7 @@ module Daru
     end
 
     # Retrive rows by positions
-    # @param [Array<Integer>] *positions positions of rows to retrive
+    # @param [Array<Integer>] positions of rows to retrive
     # @return [Daru::Vector, Daru::DataFrame] vector for single position and dataframe for multiple positions
     # @example
     #   df = Daru::DataFrame.new({
@@ -405,7 +416,7 @@ module Daru
 
     # Set rows by positions
     # @param [Array<Integer>] positions positions of rows to set
-    # @vector [Array, Daru::Vector] vector vector to be assigned
+    # @param [Array, Daru::Vector] vector vector to be assigned
     # @example
     #   df = Daru::DataFrame.new({
     #     a: [1, 2, 3],
@@ -438,7 +449,7 @@ module Daru
     end
 
     # Retrive vectors by positions
-    # @param [Array<Integer>] *positions positions of vectors to retrive
+    # @param [Array<Integer>] positions of vectors to retrive
     # @return [Daru::Vector, Daru::DataFrame] vector for single position and dataframe for multiple positions
     # @example
     #   df = Daru::DataFrame.new({
@@ -522,7 +533,7 @@ module Daru
     end
 
     def add_row row, index=nil
-      self.row[index || @size] = row
+      self.row[*(index || @size)] = row
     end
 
     def add_vector n, vector
@@ -597,7 +608,7 @@ module Daru
 
     # Returns a dataframe in which rows with any of the mentioned values
     #   are ignored.
-    # @param [Array] *values values to reject to form the new dataframe
+    # @param [Array] values to reject to form the new dataframe
     # @return [Daru::DataFrame] Data Frame with only rows which doesn't
     #   contain the mentioned values
     # @example
@@ -650,6 +661,88 @@ module Daru
       self
     end
 
+    # Rolling fillna
+    # replace all Float::NAN and NIL values with the preceeding or following value
+    #
+    # @param direction [Symbol] (:forward, :backward) whether replacement value is preceeding or following
+    #
+    # @example
+    #   df = Daru::DataFrame.new({
+    #    a: [1,    2,          3,   nil,        Float::NAN, nil, 1,   7],
+    #    b: [:a,  :b,          nil, Float::NAN, nil,        3,   5,   nil],
+    #    c: ['a',  Float::NAN, 3,   4,          3,          5,   nil, 7]
+    #   })
+    #
+    #   => #<Daru::DataFrame(8x3)>
+    #        a   b   c
+    #    0   1   a   a
+    #    1   2   b NaN
+    #    2   3 nil   3
+    #    3 nil NaN   4
+    #    4 NaN nil   3
+    #    5 nil   3   5
+    #    6   1   5 nil
+    #    7   7 nil   7
+    #
+    #   2.3.3 :068 > df.rolling_fillna(:forward)
+    #   => #<Daru::DataFrame(8x3)>
+    #        a   b   c
+    #    0   1   a   a
+    #    1   2   b   a
+    #    2   3   b   3
+    #    3   3   b   4
+    #    4   3   b   3
+    #    5   3   3   5
+    #    6   1   5   5
+    #    7   7   5   7
+    #
+    def rolling_fillna!(direction=:forward)
+      @data.each { |vec| vec.rolling_fillna!(direction) }
+    end
+
+    def rolling_fillna(direction=:forward)
+      dup.rolling_fillna!(direction)
+    end
+
+    # Return unique rows by vector specified or all vectors
+    #
+    # @param vtrs [String][Symbol] vector names(s) that should be considered
+    #
+    # @example
+    #
+    #    => #<Daru::DataFrame(6x2)>
+    #         a   b
+    #     0   1   a
+    #     1   2   b
+    #     2   3   c
+    #     3   4   d
+    #     2   3   c
+    #     3   4   f
+    #
+    #    2.3.3 :> df.unique
+    #    => #<Daru::DataFrame(5x2)>
+    #         a   b
+    #     0   1   a
+    #     1   2   b
+    #     2   3   c
+    #     3   4   d
+    #     3   4   f
+    #
+    #    2.3.3 :> df.unique(:a)
+    #    => #<Daru::DataFrame(5x2)>
+    #         a   b
+    #     0   1   a
+    #     1   2   b
+    #     2   3   c
+    #     3   4   d
+    #
+    def uniq(*vtrs)
+      vecs = vtrs.empty? ? vectors.map(&:to_s) : Array(vtrs)
+      grouped = group_by(vecs)
+      indexes = grouped.groups.values.map { |v| v[0] }.sort
+      row[*indexes]
+    end
+
     # Iterate over each index of the DataFrame.
     def each_index &block
       return to_enum(:each_index) unless block_given?
@@ -1024,9 +1117,9 @@ module Daru
       dup.tap { |df| df.keep_vector_if(&block) }
     end
 
-    # Test each row with one or more tests. Each test is a Proc with the form
-    # *Proc.new {|row| row[:age] > 0}*
-    #
+    # Test each row with one or more tests.
+    # @param tests [Proc]  Each test is a Proc with the form
+    #                      *Proc.new {|row| row[:age] > 0}*
     # The function returns an array with all errors.
     #
     # FIXME: description here is too sparse. As far as I can get,
@@ -1128,7 +1221,7 @@ module Daru
     deprecate :flawed?, :include_values?, 2016, 10
 
     # Check if any of given values occur in the data frame
-    # @param [Array] *values values to check for
+    # @param [Array] values to check for
     # @return [true, false] true if any of the given values occur in the
     #   dataframe, false otherwise
     # @example
@@ -1259,13 +1352,60 @@ module Daru
 
     alias :last :tail
 
-    # Returns a vector with sum of all vectors specified in the argument.
-    # If vecs parameter is empty, sum all numeric vector.
-    def vector_sum vecs=nil
+    # Sum all numeric/specified vectors in the DataFrame.
+    #
+    # Returns a new vector that's a containing a sum of all numeric
+    # or specified vectors of the DataFrame. By default, if the vector
+    # contains a nil, the sum is nil.
+    # With :skipnil argument set to true, nil values are assumed to be
+    # 0 (zero) and the sum vector is returned.
+    #
+    # @param args [Array] List of vectors to sum. Default is nil in which case
+    #   all numeric vectors are summed.
+    #
+    # @option opts [Boolean] :skipnil Consider nils as 0. Default is false.
+    #
+    # @return Vector with sum of all vectors specified in the argument.
+    #   If vecs parameter is empty, sum all numeric vector.
+    #
+    # @example
+    #    df = Daru::DataFrame.new({
+    #       a: [1, 2, nil],
+    #       b: [2, 1, 3],
+    #       c: [1, 1, 1]
+    #     })
+    #    => #<Daru::DataFrame(3x3)>
+    #           a   b   c
+    #       0   1   2   1
+    #       1   2   1   1
+    #       2 nil   3   1
+    #    df.vector_sum [:a, :c]
+    #    => #<Daru::Vector(3)>
+    #       0   2
+    #       1   3
+    #       2 nil
+    #    df.vector_sum
+    #    => #<Daru::Vector(3)>
+    #       0   4
+    #       1   4
+    #       2 nil
+    #    df.vector_sum skipnil: true
+    #    => #<Daru::Vector(3)>
+    #           c
+    #       0   4
+    #       1   4
+    #       2   4
+    #
+    def vector_sum(*args)
+      defaults = {vecs: nil, skipnil: false}
+      options = args.last.is_a?(::Hash) ? args.pop : {}
+      options = defaults.merge(options)
+      vecs = args[0] || options[:vecs]
+      skipnil = args[1] || options[:skipnil]
+
       vecs ||= numeric_vectors
       sum = Daru::Vector.new [0]*@size, index: @index, name: @name, dtype: @dtype
-
-      vecs.inject(sum) { |memo, n| memo + self[n] }
+      vecs.inject(sum) { |memo, n| self[n].add(memo, skipnil: skipnil) }
     end
 
     # Calculate mean of the rows of the dataframe.
@@ -1427,7 +1567,7 @@ module Daru
 
     # Reassign vectors with a new index of type Daru::Index or any of its subclasses.
     #
-    # @param [Daru::Index] idx The new index object on which the vectors are to
+    # @param new_index [Daru::Index] idx The new index object on which the vectors are to
     #   be indexed. Must of the same size as ncols.
     # @example Reassigning vectors of a DataFrame
     #   df = Daru::DataFrame.new({a: [1,2,3,4], b: [:a,:b,:c,:d], c: [11,22,33,44]})
@@ -1513,9 +1653,9 @@ module Daru
     # Sorts a dataframe (ascending/descending) in the given pripority sequence of
     # vectors, with or without a block.
     #
-    # @param order [Array] The order of vector names in which the DataFrame
+    # @param vector_order [Array] The order of vector names in which the DataFrame
     #   should be sorted.
-    # @param [Hash] opts The options to sort with.
+    # @param opts [Hash] opts The options to sort with.
     # @option opts [TrueClass,FalseClass,Array] :ascending (true) Sort in ascending
     #   or descending order. Specify Array corresponding to *order* for multiple
     #   sort orders.
@@ -1684,12 +1824,11 @@ module Daru
 
       new_fields = (@vectors.to_a + other_df.vectors.to_a)
       new_fields = ArrayHelper.recode_repeated(new_fields)
-
       DataFrame.new({}, order: new_fields).tap do |df_new|
         (0...nrows).each do |i|
           df_new.add_row row[i].to_a + other_df.row[i].to_a
         end
-
+        df_new.index = @index if @index == other_df.index
         df_new.update
       end
     end
@@ -2035,7 +2174,7 @@ module Daru
     end
 
     # Converts the specified non category type vectors to category type vectors
-    # @param [Array] *names names of non category type vectors to be converted
+    # @param [Array] names of non category type vectors to be converted
     # @return [Daru::DataFrame] data frame in which specified vectors have been
     #   converted to category type
     # @example
@@ -2126,8 +2265,88 @@ module Daru
       res
     end
 
+    # Function to use for aggregating the data.
+    #
+    # @param options [Hash] options for column, you want in resultant dataframe
+    #
+    # @return [Daru::DataFrame]
+    #
+    # @example
+    #   df = Daru::DataFrame.new(
+    #      {col: [:a, :b, :c, :d, :e], num: [52,12,07,17,01]})
+    #   => #<Daru::DataFrame(5x2)>
+    #        col num
+    #      0   a  52
+    #      1   b  12
+    #      2   c   7
+    #      3   d  17
+    #      4   e   1
+    #
+    #    df.aggregate(num_100_times: ->(df) { df.num*100 })
+    #   => #<Daru::DataFrame(5x1)>
+    #               num_100_ti
+    #             0       5200
+    #             1       1200
+    #             2        700
+    #             3       1700
+    #             4        100
+    #
+    #   When we have duplicate index :
+    #
+    #   idx = Daru::CategoricalIndex.new [:a, :b, :a, :a, :c]
+    #   df = Daru::DataFrame.new({num: [52,12,07,17,01]}, index: idx)
+    #   => #<Daru::DataFrame(5x1)>
+    #        num
+    #      a  52
+    #      b  12
+    #      a   7
+    #      a  17
+    #      c   1
+    #
+    #   df.aggregate(num: :mean)
+    #   => #<Daru::DataFrame(3x1)>
+    #                      num
+    #             a 25.3333333
+    #             b         12
+    #             c          1
+    #
+    # Note: `GroupBy` class `aggregate` method uses this `aggregate` method
+    # internally.
+    def aggregate(options={})
+      colmn_value, index_tuples = aggregated_colmn_value(options)
+      Daru::DataFrame.new(
+        colmn_value, index: index_tuples, order: options.keys
+      )
+    end
+
     private
 
+    # Do the `method` (`method` can be :sum, :mean, :std, :median, etc or
+    # lambda), on the column.
+    def apply_method_on_colmns colmn, index_tuples, method
+      rows = []
+      index_tuples.each do |indexes|
+        # If single element then also make it vector.
+        slice = Daru::Vector.new(Array(self[colmn][*indexes]))
+        case method
+        when Symbol
+          rows << (slice.is_a?(Daru::Vector) ? slice.send(method) : slice)
+        when Proc
+          rows << method.call(slice)
+        end
+      end
+      rows
+    end
+
+    def apply_method_on_df index_tuples, method
+      rows = []
+      index_tuples.each do |indexes|
+        slice = row[*indexes]
+        rows << method.call(slice)
+      end
+      rows
+    end
+
     def headers
       Daru::Index.new(Array(index.name) + @vectors.to_a)
     end
@@ -2224,9 +2443,7 @@ module Daru
         rescue IndexError
           raise IndexError, "Specified vector #{names.first} does not exist"
         end
-
         return @data[pos] if pos.is_a?(Numeric)
-
         names = pos
       end
 
@@ -2396,7 +2613,7 @@ module Daru
     end
 
     def create_vectors_index_with vectors, source
-      vectors = source.keys.sort_by(&:to_s) if vectors.nil?
+      vectors = source.keys if vectors.nil?
 
       @vectors =
         if vectors.is_a?(Index) || vectors.is_a?(MultiIndex)
@@ -2443,9 +2660,7 @@ module Daru
       @index   = Index.coerce(index || source[0].size)
       @vectors = Index.coerce(vectors)
 
-      @data = @vectors.each_with_index.map do |_vec,idx|
-        Daru::Vector.new(source[idx], index: @index, name: vectors[idx])
-      end
+      update_data source, vectors
     end
 
     def initialize_from_array_of_vectors source, vectors, index, opts
@@ -2694,6 +2909,30 @@ module Daru
       end
     end
 
+    def update_data source, vectors
+      @data = @vectors.each_with_index.map do |_vec,idx|
+        Daru::Vector.new(source[idx], index: @index, name: vectors[idx])
+      end
+    end
+
+    def aggregated_colmn_value(options)
+      colmn_value = []
+      index_tuples = Array(@index).uniq
+      options.keys.each do |vec|
+        do_this_on_vec = options[vec]
+        colmn_value << if @vectors.include?(vec)
+                         apply_method_on_colmns(
+                           vec, index_tuples, do_this_on_vec
+                         )
+                       else
+                         apply_method_on_df(
+                           index_tuples, do_this_on_vec
+                         )
+                       end
+      end
+      [colmn_value, index_tuples]
+    end
+
     # coerce ranges, integers and array in appropriate ways
     def coerce_positions *positions, size
       if positions.size == 1