Class: RailsDataExplorer::DataType::Categorical
- Inherits:
-
RailsDataExplorer::DataType
- Object
- RailsDataExplorer::DataType
- RailsDataExplorer::DataType::Categorical
- Defined in:
- lib/rails_data_explorer/data_type/categorical.rb
Overview
Responsibilities:
* Provide available charts and statistics for categorical data type.
* Provide methods for categorical data type.
Collaborators:
* DataSet
Instance Method Summary collapse
- #all_available_chart_types ⇒ Object
- #axis_tick_format(values) ⇒ Object
- #descriptive_statistics(values) ⇒ Object
-
#descriptive_statistics_table(values) ⇒ Object
Returns an object that describes a statistics table.
- #descriptive_statistics_table_horizontal(desc_stats) ⇒ Object
- #descriptive_statistics_table_vertical(desc_stats) ⇒ Object
-
#label_sorter(label_val_key, data_series, value_sorter) ⇒ Proc
A Proc that will be used by #sort.
-
#limit_distinct_values(values, max_num_vals, val_for_others = nil) ⇒ Object
Returns the top N max frequent distinct observations in values.
Methods inherited from RailsDataExplorer::DataType
Instance Method Details
#all_available_chart_types ⇒ Object
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 |
# File 'lib/rails_data_explorer/data_type/categorical.rb', line 15 def all_available_chart_types [ { chart_class: Chart::HistogramCategorical, chart_roles: [:x], dimensions_count_min: 1, dimensions_count_max: 1, }, { chart_class: Chart::PieChart, chart_roles: [:any], dimensions_count_min: 1, dimensions_count_max: 1, }, { chart_class: Chart::BoxPlotGroup, chart_roles: [:y], dimensions_count_min: 2, dimensions_count_max: 2, }, { chart_class: Chart::Scatterplot, chart_roles: [:color], dimensions_count_min: 3, }, { chart_class: Chart::ParallelCoordinates, chart_roles: [:dimension], dimensions_count_min: 3, }, { chart_class: Chart::StackedBarChartCategorical, chart_roles: [:x, :y], dimensions_count_min: 2, dimensions_count_max: 2, }, { chart_class: Chart::StackedBarChartCategoricalPercent, chart_roles: [:x, :y], dimensions_count_min: 2, dimensions_count_max: 2, }, # { # chart_class: Chart::StackedHistogramTemporal, # chart_roles: [:y], # dimensions_count_min: 2, # dimensions_count_max: 2, # }, { chart_class: Chart::ContingencyTable, chart_roles: [:any], dimensions_count_min: 2, dimensions_count_max: 2, }, { chart_class: Chart::DescriptiveStatisticsTable, chart_roles: [:any], dimensions_count_min: 1, dimensions_count_max: 1, }, { chart_class: Chart::ParallelSet, chart_roles: [:dimension], dimensions_count_min: 3, }, ].freeze end |
#axis_tick_format(values) ⇒ Object
181 182 183 |
# File 'lib/rails_data_explorer/data_type/categorical.rb', line 181 def axis_tick_format(values) %(function(d) { return d }) end |
#descriptive_statistics(values) ⇒ Object
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 |
# File 'lib/rails_data_explorer/data_type/categorical.rb', line 83 def descriptive_statistics(values) frequencies = compute_histogram(values) labels_ds = DataSeries.new('_', values.uniq) total_count = values.length ruby_formatters = { integer: Proc.new { |v| number_with_delimiter(v.round) }, percent: Proc.new { |v| number_to_percentage(v, precision: 3, significant: true, strip_insignificant_zeros: true, delimiter: ',') }, } r = frequencies.inject([]) { |m, (k,v)| m << { label: "#{ k.to_s }_count", value: v, ruby_formatter: ruby_formatters[:integer] } m << { label: "#{ k.to_s }_percent", value: (v / total_count.to_f) * 100, ruby_formatter: ruby_formatters[:percent] } m }.sort( &labels_ds.label_sorter( :label, lambda { |a,b| b[:value] <=> a[:value] } ) ) r.insert(0, { label: '[Total]_count', value: total_count, ruby_formatter: ruby_formatters[:integer] }) r.insert(0, { label: '[Total]_percent', value: 100, ruby_formatter: ruby_formatters[:percent] }) r end |
#descriptive_statistics_table(values) ⇒ Object
Returns an object that describes a statistics table.
107 108 109 110 111 112 113 114 |
# File 'lib/rails_data_explorer/data_type/categorical.rb', line 107 def descriptive_statistics_table(values) desc_stats = descriptive_statistics(values) if desc_stats.length < DataSeries.many_uniq_vals_threshold descriptive_statistics_table_horizontal(desc_stats) else descriptive_statistics_table_vertical(desc_stats) end end |
#descriptive_statistics_table_horizontal(desc_stats) ⇒ Object
116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 |
# File 'lib/rails_data_explorer/data_type/categorical.rb', line 116 def descriptive_statistics_table_horizontal(desc_stats) labels = desc_stats.map { |e| e[:label].gsub(/_count|_percent/, '') }.uniq table = Utils::RdeTable.new([]) table.rows << Utils::RdeTableRow.new( :tr, labels.map { |label| Utils::RdeTableCell.new(:th, label, ruby_formatter: Proc.new { |e| e }, css_class: 'rde-cell-label') }, css_class: 'rde-column_header' ) table.rows << Utils::RdeTableRow.new( :tr, labels.map { |label| stat = desc_stats.detect { |e| "#{ label }_count" == e[:label] } Utils::RdeTableCell.new(:td, stat[:value], ruby_formatter: stat[:ruby_formatter], css_class: 'rde-cell-value') }, css_class: 'rde-data_row' ) table.rows << Utils::RdeTableRow.new( :tr, labels.map { |label| stat = desc_stats.detect { |e| "#{ label }_percent" == e[:label] } Utils::RdeTableCell.new(:td, stat[:value], ruby_formatter: stat[:ruby_formatter], css_class: 'rde-cell-value') }, css_class: 'rde-data_row' ) table end |
#descriptive_statistics_table_vertical(desc_stats) ⇒ Object
145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 |
# File 'lib/rails_data_explorer/data_type/categorical.rb', line 145 def descriptive_statistics_table_vertical(desc_stats) labels = desc_stats.map { |e| e[:label].gsub(/_count|_percent/, '') }.uniq table = Utils::RdeTable.new([]) table.rows << Utils::RdeTableRow.new( :tr, %w[Value Count Percent].map { |label| Utils::RdeTableCell.new(:th, label, css_class: 'rde-cell-label') }, css_class: 'rde-column_header', ) labels.each { |label| count_stat = desc_stats.detect { |e| "#{ label }_count" == e[:label] } percent_stat = desc_stats.detect { |e| "#{ label }_percent" == e[:label] } table.rows << Utils::RdeTableRow.new( :tr, [ Utils::RdeTableCell.new(:td, label, css_class: 'rde-cell-value'), Utils::RdeTableCell.new( :td, count_stat[:value], ruby_formatter: count_stat[:ruby_formatter], css_class: 'rde-cell-value' ), Utils::RdeTableCell.new( :td, percent_stat[:value], ruby_formatter: percent_stat[:ruby_formatter], css_class: 'rde-cell-value' ), ], css_class: 'rde-data_row', ) } table end |
#label_sorter(label_val_key, data_series, value_sorter) ⇒ Proc
Returns a Proc that will be used by #sort.
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 |
# File 'lib/rails_data_explorer/data_type/categorical.rb', line 189 def label_sorter(label_val_key, data_series, value_sorter) if data_series.uniq_vals.any? { |e| e.to_s =~ /^[\+\-]?\d+/ } # Sort numerical categories by key ASC # This lambda can be used in conjunction with `#sort`. # It returns -1, 0, or 1 lambda { |a,b| number_and_full_string_extractor = lambda { |val| str = label_val_key ? val[label_val_key] : val number = str.gsub(/^[^\d\+\-]*/, '') # remove non-digit leading chars .gsub(',', '') # remove delimiter commas, they throw off to_f parsing if '' != number # label contains digits number = number.to_f number += 1 if str =~ /^>/ # increase highest threshold by one for proper sorting number -= 1 if str =~ /^</ # decrease lowest threshold by one for proper sorting else # label doesn't contain digits, set to nil to sort at end number = nil end [number, str] } a_num, a_str = number_and_full_string_extractor.call(a) b_num, b_str = number_and_full_string_extractor.call(b) if a_num && b_num # Both numbers are present, compare them [a_num, a_str] <=> [b_num, b_str] elsif a_num # a_num is present, b_num isn't. Sort a before b -1 else # a_num is not present, b_num is, Sort a after b 1 end } else # Use provided value sorter value_sorter end end |
#limit_distinct_values(values, max_num_vals, val_for_others = nil) ⇒ Object
Returns the top N max frequent distinct observations in values. Groups less frequent observations under val_for_others.
234 235 236 237 238 239 240 241 242 243 244 245 246 247 |
# File 'lib/rails_data_explorer/data_type/categorical.rb', line 234 def limit_distinct_values(values, max_num_vals, val_for_others = nil) distinct_values = values.uniq # Return values if they already have lte max_num_vals distinct observations return values if distinct_values.length <= max_num_vals val_for_others ||= '[Other]' frequencies = compute_histogram(values) top_vals = frequencies.to_a.sort { |a,b| # a = [value, frequency] # Sort by frequency DESC, value ASC [b.last, a.first] <=> [a.last, b.first] }.first(max_num_vals - 1).map { |e| e.first } values.map { |e| top_vals.include?(e) ? e : val_for_others } end |