rover-df 0.1.1 → 0.2.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: b003d311b623fdd38fee4c6fa76129ad4bba042e8193c1872928cb05085daad3
-  data.tar.gz: d0c8c04b2a8aec3ea5b7616cbcda61f03a12c96fc8f9a0c7aa29fc898948b759
+  metadata.gz: 0a96f5fa610232beb50eb2d22cd87b52627dfda44139e618fb0fa5d82f4d90fa
+  data.tar.gz: 899ae175705f7195a994c0bfd19fd021334404d07440f05153c08be5fa08c11b
 SHA512:
-  metadata.gz: 8033d8ae9e5fb8c8e767ba68897d37745cc5d35a7a82cb2847e2b1d2c3adf8eeb406914cd47949d8f4c3f21307617ab550f435e7a5c257fe1187ed47dd943829
-  data.tar.gz: acfdca4ad081e2722c4b5269824de123d26aadab28532837d2bfc717c2ca263f73dc4335963beae654b24d4c17cebbefdcce761a946f391340a306c9ca2a8c9b
+  metadata.gz: dc61f769ce5bac5fcda65dcc25fb940f7bf9f81ac168029748b5ccf8bfb896008f0836f9523ca3785e22782dff3f4d8dccc19ae823f944391c47df41fd88b3ec
+  data.tar.gz: 5a0c7e5e77233ef8bf7b4df23d711ec22a2e85165715ca330344a0b6cd5bef99e571256496e9c6f6037bf0fc0c771b4e951d91f0a85c17211127649d6e246878

data/CHANGELOG.md

@@ -1,3 +1,33 @@
+## 0.2.4 (2021-06-03)
+
+- Added grouping for `std` and `var`
+- Fixed `==` for data frames
+- Fixed error with `first` and `last` for data frames
+- Fixed error with `last` when vector size is smaller than `n`
+
+## 0.2.3 (2021-02-08)
+
+- Added `select`, `reject`, and `map!` methods to vectors
+
+## 0.2.2 (2021-01-01)
+
+- Added line, pie, area, and bar charts
+- Added `|` and `^` for vectors
+- Fixed typecasting with `map`
+
+## 0.2.1 (2020-11-23)
+
+- Added `plot` method to data frames
+- Improved error message when too few headers
+
+## 0.2.0 (2020-08-17)
+
+- Added `numeric?` and `zip` methods to vectors
+- Changed group calculations to return a data frame instead of a hash
+- Changed `each_row` to return enumerator
+- Improved inspect
+- Fixed `any?`, `all?`, and `uniq` for boolean vectors
+
 ## 0.1.1 (2020-06-10)
 
 - Added methods and options for types

data/LICENSE.txt

@@ -1,4 +1,4 @@
-Copyright (c) 2020 Andrew Kane
+Copyright (c) 2020-2021 Andrew Kane
 
 MIT License
 

data/README.md

@@ -4,7 +4,9 @@ Simple, powerful data frames for Ruby
 
 :mountain: Designed for data exploration and machine learning, and powered by [Numo](https://github.com/ruby-numo/numo-narray)
 
-[![Build Status](https://travis-ci.org/ankane/rover.svg?branch=master)](https://travis-ci.org/ankane/rover)
+:evergreen_tree: Uses [Vega](https://github.com/ankane/vega) for visualization
+
+[![Build Status](https://github.com/ankane/rover/workflows/build/badge.svg?branch=master)](https://github.com/ankane/rover/actions)
 
 ## Installation
 
@@ -121,30 +123,53 @@ df[1..3]
 df[[1, 4, 5]]
 ```
 
+Iterate over rows
+
+```ruby
+df.each_row { |row| ... }
+```
+
+Iterate over a column
+
+```ruby
+df[:a].each { |item| ... }
+# or
+df[:a].each_with_index { |item, index| ... }
+```
+
 ## Filtering
 
 Filter on a condition
 
 ```ruby
+df[df[:a] == 100]
+df[df[:a] != 100]
 df[df[:a] > 100]
+df[df[:a] >= 100]
+df[df[:a] < 100]
+df[df[:a] <= 100]
 ```
 
-And
+In
 
 ```ruby
-df[df[:a] > 100 & df[:b] == "one"]
+df[df[:a].in?([1, 2, 3])]
+df[df[:a].in?(1..3)]
+df[df[:a].in?(["a", "b", "c"])]
 ```
 
-Or
+Not in
 
 ```ruby
-df[df[:a] > 100 | df[:b] == "one"]
+df[!df[:a].in?([1, 2, 3])]
 ```
 
-Not
+And, or, and exclusive or
 
 ```ruby
-df[df[:a] != 100]
+df[(df[:a] > 100) & (df[:b] == "one")] # and
+df[(df[:a] > 100) | (df[:b] == "one")] # or
+df[(df[:a] > 100) ^ (df[:b] == "one")] # xor
 ```
 
 ## Operations
@@ -170,6 +195,8 @@ df[:a].median
 df[:a].percentile(90)
 df[:a].min
 df[:a].max
+df[:a].std
+df[:a].var
 ```
 
 Count occurrences
@@ -204,6 +231,26 @@ Multiple groups
 df.group([:a, :b]).count
 ```
 
+## Visualization
+
+Add [Vega](https://github.com/ankane/vega) to your application’s Gemfile:
+
+```ruby
+gem 'vega'
+```
+
+And use:
+
+```ruby
+df.plot(:a, :b)
+```
+
+Specify the chart type (`line`, `pie`, `column`, `bar`, `area`, or `scatter`)
+
+```ruby
+df.plot(:a, :b, type: "pie")
+```
+
 ## Updating Data
 
 Add a new column
@@ -228,6 +275,14 @@ df[:a][0..2] = 1
 df[:a][0..2] = [1, 2, 3]
 ```
 
+Update all elements
+
+```ruby
+df[:a] = df[:a].map { |v| v.gsub("a", "b") }
+# or
+df[:a].map! { |v| v.gsub("a", "b") }
+```
+
 Update elements matching a condition
 
 ```ruby
@@ -340,7 +395,7 @@ df.to_csv
 
 ## Types
 
-Pass column types when creating a data frame
+You can specify column types when creating a data frame
 
 ```ruby
 Rover::DataFrame.new(data, types: {"a" => :int, "b" => :float})

data/lib/rover.rb

@@ -11,12 +11,12 @@ module Rover
   class << self
     def read_csv(path, types: nil, **options)
       require "csv"
-      csv_to_df(CSV.read(path, **csv_options(options)), types: types)
+      csv_to_df(CSV.read(path, **csv_options(options)), types: types, headers: options[:headers])
     end
 
     def parse_csv(str, types: nil, **options)
       require "csv"
-      csv_to_df(CSV.parse(str, **csv_options(options)), types: types)
+      csv_to_df(CSV.parse(str, **csv_options(options)), types: types, headers: options[:headers])
     end
 
     private
@@ -28,7 +28,11 @@ module Rover
       options
     end
 
-    def csv_to_df(table, types: nil)
+    def csv_to_df(table, types: nil, headers: nil)
+      if headers && headers.size < table.headers.size
+        raise ArgumentError, "Expected #{table.headers.size} headers, got #{headers.size}"
+      end
+
       table.by_col!
       data = {}
       table.each do |k, v|

data/lib/rover/data_frame.rb

@@ -72,6 +72,7 @@ module Rover
         # multiple columns
         df = DataFrame.new
         where.each do |k|
+          check_column(k, true)
           df[k] = @vectors[k]
         end
         df
@@ -81,8 +82,9 @@ module Rover
       end
     end
 
-    # return each row as a hash
     def each_row
+      return enum_for(:each_row) unless block_given?
+
       size.times do |i|
         yield @vectors.map { |k, v| [k, v[i]] }.to_h
       end
@@ -161,7 +163,7 @@ module Rover
       last(n)
     end
 
-    def first(n = nil)
+    def first(n = 1)
       new_vectors = {}
       @vectors.each do |k, v|
         new_vectors[k] = v.first(n)
@@ -169,7 +171,7 @@ module Rover
       DataFrame.new(new_vectors)
     end
 
-    def last(n = nil)
+    def last(n = 1)
       new_vectors = {}
       @vectors.each do |k, v|
         new_vectors[k] = v.last(n)
@@ -236,7 +238,12 @@ module Rover
     # for IRuby
     def to_html
       require "iruby"
-      IRuby::HTML.table(to_h)
+      if size > 7
+        # pass 8 rows so maxrows is applied
+        IRuby::HTML.table((self[0..4] + self[-4..-1]).to_h, maxrows: 7)
+      else
+        IRuby::HTML.table(to_h)
+      end
     end
 
     # TODO handle long text better
@@ -247,18 +254,19 @@ module Rover
       line_start = 0
       spaces = 2
 
+      summarize = size >= 30
+
       @vectors.each do |k, v|
-        v = v.first(5).to_a
+        v = summarize ? v.first(5).to_a + ["..."] + v.last(5).to_a : v.to_a
         width = ([k] + v).map(&:to_s).map(&:size).max
         width = 3 if width < 3
 
         if lines.empty? || lines[-2].map { |l| l.size + spaces }.sum + width > 120
           line_start = lines.size
           lines << []
-          [size, 5].min.times do |i|
+          v.size.times do |i|
             lines << []
           end
-          lines << [] if size > 5
           lines << []
         end
 
@@ -266,7 +274,6 @@ module Rover
         v.each_with_index do |v2, i|
           lines[line_start + 1 + i] << "%#{width}s" % v2.to_s
         end
-        lines[line_start + 6] << "%#{width}s" % "..." if size > 5
       end
 
       lines.pop
@@ -290,11 +297,11 @@ module Rover
       dup.sort_by!(&block)
     end
 
-    def group(columns)
-      Group.new(self, columns)
+    def group(*columns)
+      Group.new(self, columns.flatten)
     end
 
-    [:max, :min, :median, :mean, :percentile, :sum].each do |name|
+    [:max, :min, :median, :mean, :percentile, :sum, :std, :var].each do |name|
       define_method(name) do |column, *args|
         check_column(column)
         self[column].send(name, *args)
@@ -353,7 +360,89 @@ module Rover
     def ==(other)
       size == other.size &&
       keys == other.keys &&
-      keys.all? { |k| self[k] == other[k] }
+      keys.all? { |k| self[k].to_numo == other[k].to_numo }
+    end
+
+    def plot(x = nil, y = nil, type: nil)
+      require "vega"
+
+      raise ArgumentError, "Must specify columns" if keys.size != 2 && (!x || !y)
+      x ||= keys[0]
+      y ||= keys[1]
+      type ||= begin
+        if self[x].numeric? && self[y].numeric?
+          "scatter"
+        elsif types[x] == :object && self[y].numeric?
+          "column"
+        else
+          raise "Cannot determine type. Use the type option."
+        end
+      end
+      data = self[[x, y]]
+
+      case type
+      when "line", "area"
+        x_type =
+          if data[x].numeric?
+            "quantitative"
+          elsif data[x].all? { |v| v.is_a?(Date) || v.is_a?(Time) }
+            "temporal"
+          else
+            "nominal"
+          end
+
+        scale = x_type == "temporal" ? {type: "utc"} : {}
+
+        Vega.lite
+          .data(data)
+          .mark(type: type, tooltip: true, interpolate: "cardinal", point: {size: 60})
+          .encoding(
+            x: {field: x, type: x_type, scale: scale},
+            y: {field: y, type: "quantitative"}
+          )
+          .config(axis: {labelFontSize: 12})
+      when "pie"
+        Vega.lite
+          .data(data)
+          .mark(type: "arc", tooltip: true)
+          .encoding(
+            color: {field: x, type: "nominal", sort: "none", axis: {title: nil}, legend: {labelFontSize: 12}},
+            theta: {field: y, type: "quantitative"}
+          )
+          .view(stroke: nil)
+      when "column"
+        Vega.lite
+          .data(data)
+          .mark(type: "bar", tooltip: true)
+          .encoding(
+            # TODO determine label angle
+            x: {field: x, type: "nominal", sort: "none", axis: {labelAngle: 0}},
+            y: {field: y, type: "quantitative"}
+          )
+          .config(axis: {labelFontSize: 12})
+      when "bar"
+        Vega.lite
+          .data(data)
+          .mark(type: "bar", tooltip: true)
+          .encoding(
+            # TODO determine label angle
+            y: {field: x, type: "nominal", sort: "none", axis: {labelAngle: 0}},
+            x: {field: y, type: "quantitative"}
+          )
+          .config(axis: {labelFontSize: 12})
+      when "scatter"
+        Vega.lite
+          .data(data)
+          .mark(type: "circle", tooltip: true)
+          .encoding(
+            x: {field: x, type: "quantitative", scale: {zero: false}},
+            y: {field: y, type: "quantitative", scale: {zero: false}},
+            size: {value: 60}
+          )
+          .config(axis: {labelFontSize: 12})
+      else
+        raise ArgumentError, "Invalid type: #{type}"
+      end
     end
 
     private
@@ -418,8 +507,20 @@ module Rover
       raise ArgumentError, "Missing keys: #{missing_keys.join(", ")}" if missing_keys.any?
     end
 
-    def check_column(key)
-      raise ArgumentError, "Missing column: #{key}" unless include?(key)
+    # TODO in 0.3.0
+    # always use did_you_mean
+    def check_column(key, did_you_mean = false)
+      unless include?(key)
+        if did_you_mean
+          if RUBY_VERSION.to_f >= 2.6
+            raise KeyError.new("Missing column: #{key}", receiver: self, key: key)
+          else
+            raise KeyError.new("Missing column: #{key}")
+          end
+        else
+          raise ArgumentError, "Missing column: #{key}"
+        end
+      end
     end
 
     def to_vector(v, size: nil, type: nil)

data/lib/rover/group.rb

@@ -2,16 +2,23 @@ module Rover
   class Group
     def initialize(df, columns)
       @df = df
-      @columns = Array(columns)
+      @columns = columns
     end
 
-    [:count, :max, :min, :mean, :median, :percentile, :sum].each do |name|
+    def group(*columns)
+      Group.new(@df, @columns + columns.flatten)
+    end
+
+    [:count, :max, :min, :mean, :median, :percentile, :sum, :std, :var].each do |name|
       define_method(name) do |*args|
-        result = {}
+        n = [name, args.first].compact.join("_")
+
+        rows = []
         grouped_dfs.each do |k, df|
-          result[k] = df.send(name, *args)
+          rows << k.merge(n => df.send(name, *args))
         end
-        result
+
+        DataFrame.new(rows)
       end
     end
 
@@ -26,21 +33,15 @@ module Rover
         raise ArgumentError, "Missing keys: #{missing_keys.join(", ")}" if missing_keys.any?
 
         groups = Hash.new { |hash, key| hash[key] = [] }
-        if @columns.size == 1
-          @df[@columns.first].each_with_index do |v, i|
-            groups[v] << i
-          end
-        else
-          i = 0
-          @df.each_row do |row|
-            groups[@columns.map { |c| row[c] }] << i
-            i += 1
-          end
+        i = 0
+        @df.each_row do |row|
+          groups[row.slice(*@columns)] << i
+          i += 1
         end
 
         result = {}
-        groups.each do |k, indexes|
-          result[k] = @df[indexes]
+        groups.keys.each do |k|
+          result[k] = @df[groups[k]]
         end
         result
       end

data/lib/rover/vector.rb

@@ -44,6 +44,10 @@ module Rover
       a
     end
 
+    def numeric?
+      ![:object, :bool].include?(type)
+    end
+
     def size
       @data.size
     end
@@ -51,7 +55,7 @@ module Rover
     alias_method :count, :size
 
     def uniq
-      Vector.new(@data.to_a.uniq)
+      Vector.new(to_a.uniq)
     end
 
     def missing
@@ -87,11 +91,11 @@ module Rover
       @data[k] = v
     end
 
-    %w(+ - * / % ** &).each do |op|
+    %w(+ - * / % ** & | ^).each do |op|
       define_method(op) do |other|
         other = other.to_numo if other.is_a?(Vector)
         # TODO better logic
-        if @data.is_a?(Numo::RObject)
+        if @data.is_a?(Numo::RObject) && !other.is_a?(Numo::RObject)
           map { |v| v.send(op, other) }
         else
           Vector.new(@data.send(op, other))
@@ -157,9 +161,22 @@ module Rover
     end
 
     def map(&block)
-      mapped = @data.map(&block)
-      mapped = mapped.to_a if mapped.is_a?(Numo::RObject) # re-evaluate cast
-      Vector.new(mapped)
+      # convert to Ruby first to cast properly
+      # https://github.com/ruby-numo/numo-narray/issues/181
+      Vector.new(@data.to_a.map(&block))
+    end
+
+    def map!(&block)
+      @data = cast_data(@data.to_a.map(&block))
+      self
+    end
+
+    def select(&block)
+      Vector.new(@data.to_a.select(&block))
+    end
+
+    def reject(&block)
+      Vector.new(@data.to_a.reject(&block))
     end
 
     def tally
@@ -226,11 +243,15 @@ module Rover
     end
 
     def all?(&block)
-      @data.to_a.all?(&block)
+      to_a.all?(&block)
     end
 
     def any?(&block)
-      @data.to_a.any?(&block)
+      to_a.any?(&block)
+    end
+
+    def zip(other, &block)
+      to_a.zip(other.to_a, &block)
     end
 
     def first(n = 1)
@@ -242,7 +263,11 @@ module Rover
     end
 
     def last(n = 1)
-      Vector.new(@data[-n..-1])
+      if n >= size
+        Vector.new(@data)
+      else
+        Vector.new(@data[-n..-1])
+      end
     end
 
     def take(n)
@@ -298,7 +323,12 @@ module Rover
     # for IRuby
     def to_html
       require "iruby"
-      IRuby::HTML.table(to_a)
+      if size > 7
+        # pass 8 rows so maxrows is applied
+        IRuby::HTML.table(first(4).to_a + last(4).to_a, maxrows: 7)
+      else
+        IRuby::HTML.table(to_a)
+      end
     end
 
     private

data/lib/rover/version.rb

@@ -1,3 +1,3 @@
 module Rover
-  VERSION = "0.1.1"
+  VERSION = "0.2.4"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rover-df
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.2.4
 platform: ruby
 authors:
 - Andrew Kane
-autorequire: 
+autorequire:
 bindir: bin
 cert_chain: []
-date: 2020-06-10 00:00:00.000000000 Z
+date: 2021-06-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: numo-narray
@@ -16,100 +16,16 @@ dependencies:
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.9.1.7
+        version: 0.9.1.9
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.9.1.7
-- !ruby/object:Gem::Dependency
-  name: bundler
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-- !ruby/object:Gem::Dependency
-  name: rake
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-- !ruby/object:Gem::Dependency
-  name: minitest
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '5'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '5'
-- !ruby/object:Gem::Dependency
-  name: activerecord
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '5'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '5'
-- !ruby/object:Gem::Dependency
-  name: sqlite3
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-- !ruby/object:Gem::Dependency
-  name: iruby
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-description: 
-email: andrew@chartkick.com
+        version: 0.9.1.9
+description:
+email: andrew@ankane.org
 executables: []
 extensions: []
 extra_rdoc_files: []
@@ -127,7 +43,7 @@ homepage: https://github.com/ankane/rover
 licenses:
 - MIT
 metadata: {}
-post_install_message: 
+post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -142,8 +58,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.1.2
-signing_key: 
+rubygems_version: 3.2.3
+signing_key:
 specification_version: 4
 summary: Simple, powerful data frames for Ruby
 test_files: []