Class: Rust::DataFrame

Inherits:
RustDatatype show all
Defined in:
lib/rust/core/types/dataframe.rb

Class Method Summary collapse

Instance Method Summary collapse

Methods inherited from RustDatatype

#r_hash, #r_mirror, #r_mirror_to

Constructor Details

#initialize(labels_or_data) ⇒ DataFrame

Returns a new instance of DataFrame.



22
23
24
25
26
27
28
29
30
31
32
33
34
35
 
# File 'lib/rust/core/types/dataframe.rb', line 22

def initialize(labels_or_data)
    @data = {}
    
    if labels_or_data.is_a? Array
        @labels = labels_or_data.map { |l| l.to_s }
        @labels.each { |label| @data[label] = [] }
    elsif labels_or_data.is_a? Hash
        @labels = labels_or_data.keys.map { |l| l.to_s }
        
        labels_or_data.each do |key, value|
            @data[key.to_s] = value.clone
        end
    end
end

Class Method Details

.can_pull?(type, klass) ⇒ Boolean

Returns:

  • (Boolean) 


5
6
7
 
# File 'lib/rust/core/types/dataframe.rb', line 5

def self.can_pull?(type, klass)
    return [klass].flatten.include?("data.frame")
end

.pull_priority ⇒ Object 



9
10
11
 
# File 'lib/rust/core/types/dataframe.rb', line 9

def self.pull_priority
    1
end

.pull_variable(variable, type, klass) ⇒ Object 



13
14
15
16
17
18
19
20
 
# File 'lib/rust/core/types/dataframe.rb', line 13

def self.pull_variable(variable, type, klass)
    hash = {}
    colnames = Rust["colnames(#{variable})"]
    colnames.each do |col|
        hash[col] = Rust["#{variable}$\"#{col}\""]
    end
    return DataFrame.new(hash)
end

Instance Method Details

#[](rows, cols = nil) ⇒ Object 



67
68
69
70
71
72
73
74
75
76
77
78
79
80
 
# File 'lib/rust/core/types/dataframe.rb', line 67

def [](rows, cols=nil)
    raise "You must specify either rows or columns to select" if !rows && !cols
    result = self
    if rows && (rows.is_a?(Range) || rows.is_a?(Array))
        result = result.select_rows { |row, i| rows.include?(i) }
    end
    
    if cols && cols.is_a?(Array)
        cols = cols.map { |c| c.to_s }
        result = result.select_columns(cols)
    end
    
    return result
end

#add_column(name, values = nil) ⇒ Object 



204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
 
# File 'lib/rust/core/types/dataframe.rb', line 204

def add_column(name, values=nil)
    raise "Column already exists" if @labels.include?(name)
    raise "Values or block required" if !values && !block_given?
    raise "Number of values not matching" if values && values.size != self.rows
    
    @labels << name
    if values
        @data[name] = values.clone
    else
        @data[name] = []
        self.each_with_index do |row, i|
            @data[name][i] = yield row
        end
    end
end

#add_row(row) ⇒ Object Also known as: << 



181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
 
# File 'lib/rust/core/types/dataframe.rb', line 181

def add_row(row)
    if row.is_a?(Array)
        raise "Expected an array of size #{@data.size}" unless row.size == @data.size
        
        @labels.each_with_index do |label, i|
            @data[label] << row[i]
        end
        
        return true
    elsif row.is_a?(Hash)
        raise "Expected a hash with the following keys: #{@data.keys}" unless row.keys.map { |l| l.to_s }.sort == @data.keys.sort
        
        row.each do |key, value|
            @data[key.to_s] << value
        end
        
        return true
    else
        raise TypeError, "Expected an Array or a Hash"
    end
end

#aggregate(by, **aggregators) ⇒ Object

Raises:

  • (TypeError) 


379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
 
# File 'lib/rust/core/types/dataframe.rb', line 379

def aggregate(by, **aggregators)
    raise TypeError, "Expected a string" unless by.is_a?(String)
    raise TypeError, "All the aggregators should be procs" unless aggregators.values.all? { |v| v.is_a?(Proc) }
    raise "Expected a block for default aggregator" unless block_given?
    
    aggregators = aggregators.map { |label, callable| [label.to_s, callable] }.to_h
    
    sorted = self.sort_by(by)
    
    current_value = nil
    partials = []
    partial = nil
    sorted.column(by).each_with_index do |value, index|
        if current_value != value
            current_value = value
            partials << partial if partial
            partial = Rust::DataFrame.new(self.column_names)
        end
        partial << sorted.fast_row(index)
    end
    partials << partial
    
    result = Rust::DataFrame.new(self.column_names)
    partials.each do |partial|
        aggregated_row = {}
        aggregated_row[by] = partial.column(by)[0]
        (self.column_names - [by]).each do |column|
            if aggregators[column]
                aggregated_row[column] = aggregators[column].call(partial.column(column))
            else
                aggregated_row[column] = yield partial.column(column)
            end
        end
        
        result << aggregated_row
    end
    
    return result
end

#bind_columns(dataframe) ⇒ Object Also known as: cbind 



482
483
484
485
486
 
# File 'lib/rust/core/types/dataframe.rb', line 482

def bind_columns(dataframe)
    result = self.clone
    result.bind_columns!(dataframe)
    return result
end

#bind_columns!(dataframe) ⇒ Object Also known as: cbind!

Raises:

  • (TypeError) 


462
463
464
465
466
467
468
469
470
471
472
 
# File 'lib/rust/core/types/dataframe.rb', line 462

def bind_columns!(dataframe)
    raise TypeError, "DataFrame expected" unless dataframe.is_a?(DataFrame)
    raise "The number of rows are not compatible" if self.rows != dataframe.rows
    raise "The dataset would override some columns" if (self.column_names & dataframe.column_names).size > 0
    
    dataframe.column_names.each do |column_name|
        self.add_column(column_name, dataframe.column(column_name))
    end
    
    return true
end

#bind_rows(dataframe) ⇒ Object Also known as: rbind 



475
476
477
478
479
 
# File 'lib/rust/core/types/dataframe.rb', line 475

def bind_rows(dataframe)
    result = self.clone
    result.bind_rows!(dataframe)
    return result
end

#bind_rows!(dataframe) ⇒ Object Also known as: rbind!

Raises:

  • (TypeError) 


450
451
452
453
454
455
456
457
458
459
 
# File 'lib/rust/core/types/dataframe.rb', line 450

def bind_rows!(dataframe)
    raise TypeError, "DataFrame expected" unless dataframe.is_a?(DataFrame)
    raise "The columns are not compatible: #{self.column_names - dataframe.column_names} - #{dataframe.column_names - self.column_names}" unless (self.column_names & dataframe.column_names).size == self.columns
    
    dataframe.each do |row|
        self << row
    end
    
    return true
end

#clone ⇒ Object 



489
490
491
 
# File 'lib/rust/core/types/dataframe.rb', line 489

def clone
    DataFrame.new(@data)
end

#column(name) ⇒ Object Also known as: | 



82
83
84
 
# File 'lib/rust/core/types/dataframe.rb', line 82

def column(name)
    return @data[name]
end

#column_names ⇒ Object Also known as: colnames 



168
169
170
 
# File 'lib/rust/core/types/dataframe.rb', line 168

def column_names
    return @labels.map { |k| k.to_s }
end

#columns ⇒ Object 



177
178
179
 
# File 'lib/rust/core/types/dataframe.rb', line 177

def columns
    @labels.size
end

#delete_column(column) ⇒ Object 



129
130
131
132
 
# File 'lib/rust/core/types/dataframe.rb', line 129

def delete_column(column)
    @labels.delete(column)
    @data.delete(column)
end

#delete_row(i) ⇒ Object 



134
135
136
137
138
 
# File 'lib/rust/core/types/dataframe.rb', line 134

def delete_row(i)
    @data.each do |label, column|
        column.delete_at(i)
    end
end

#each ⇒ Object 



220
221
222
223
224
225
226
 
# File 'lib/rust/core/types/dataframe.rb', line 220

def each
    self.each_with_index do |element, i|
        yield element
    end
    
    return self
end

#each_with_index ⇒ Object 



236
237
238
239
240
241
242
243
244
245
246
247
 
# File 'lib/rust/core/types/dataframe.rb', line 236

def each_with_index
    for i in 0...self.rows
        element = {}
        @labels.each do |label|
            element[label] = @data[label][i]
        end
        
        yield element, i
    end
    
    return self
end

#fast_each ⇒ Object 



228
229
230
231
232
233
234
 
# File 'lib/rust/core/types/dataframe.rb', line 228

def fast_each
    self.fast_each_with_index do |element, i|
        yield element
    end
    
    return self
end

#fast_each_with_index ⇒ Object 



249
250
251
252
253
254
255
256
257
258
259
260
 
# File 'lib/rust/core/types/dataframe.rb', line 249

def fast_each_with_index
    for i in 0...self.rows
        element = []
        @labels.each do |label|
            element << @data[label][i]
        end
        
        yield element, i
    end
    
    return self
end

#fast_row(i) ⇒ Object 



45
46
47
48
49
50
51
 
# File 'lib/rust/core/types/dataframe.rb', line 45

def fast_row(i)
    if i < 0 || i >= self.rows
        return nil
    else
        return @labels.map { |label| @data[label][i] }
    end
end

#has_row? ⇒ Boolean

Returns:

  • (Boolean) 


107
108
109
110
111
112
 
# File 'lib/rust/core/types/dataframe.rb', line 107

def has_row?
    self.each_with_index do |row, i|
        return true if yield row, i
    end
    return false
end

#head(n = 10) ⇒ Object 



305
306
307
308
309
310
311
 
# File 'lib/rust/core/types/dataframe.rb', line 305

def head(n=10)
    result = DataFrame.new(self.column_names)
    self.each_with_index do |row, i|
        result << row if i < n
    end
    return result
end

#inspect ⇒ Object 



284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
 
# File 'lib/rust/core/types/dataframe.rb', line 284

def inspect
    separator = " | "
    col_widths = self.column_names.map { |colname| [colname, ([colname.length] + @data[colname].map { |e| e.inspect.length }).max] }.to_h
    col_widths[:rowscol] = (self.rows - 1).inspect.length + 3
    
    result = ""
    result << "-" * (col_widths.values.sum + ((col_widths.size - 1) * separator.length)) + "\n"
    result << (" " * col_widths[:rowscol]) + self.column_names.map { |colname| (" " * (col_widths[colname] - colname.length)) + colname }.join(separator) + "\n"
    result << "-" * (col_widths.values.sum + ((col_widths.size - 1) * separator.length)) + "\n"
    self.each_with_index do |row, i|
        index_part = "[" + (" " * (col_widths[:rowscol] - i.inspect.length - 3)) + "#{i}] "
        row_part   = row.map { |colname, value| (" " * (col_widths[colname] - value.inspect.length)) + value.inspect }.join(separator)
        
        result << index_part + row_part + "\n"
    end
    
    result << "-" * (col_widths.values.sum + ((col_widths.size - 1) * separator.length))
    
    return result
end

#load_in_r_as(variable_name) ⇒ Object 



262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
 
# File 'lib/rust/core/types/dataframe.rb', line 262

def load_in_r_as(variable_name)
    command = []
    
    command << "#{variable_name} <- data.frame()"
    row_index = 1
    self.each do |row|
        command << "#{variable_name}[#{row_index.to_R}, #{row.keys.to_R}] <- #{row.values.to_R}"
        
        row_index += 1
    end
    
    self.column_names.each do |name|
        column = self.column(name)
        
        if column.is_a?(Factor)
            command << "#{variable_name}[,#{name.to_R}] <- factor(#{variable_name}[,#{name.to_R}], labels=#{column.levels.to_R})"
        end
    end
    
    Rust._eval_big(command)
end

#merge(other, by, first_alias = "x", second_alias = "y") ⇒ Object

Raises:

  • (TypeError) 


313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
 
# File 'lib/rust/core/types/dataframe.rb', line 313

def merge(other, by, first_alias = "x", second_alias = "y")
    raise TypeError, "Expected Rust::DataFrame" unless other.is_a?(DataFrame)
    raise TypeError, "Expected list of strings" if !by.is_a?(Array) || !by.all? { |e| e.is_a?(String) }
    raise "This dataset should have all the columns in #{by}" unless (by & self.column_names).size == by.size
    raise "The passed dataset should have all the columns in #{by}" unless (by & other.column_names).size == by.size
    
    if first_alias == second_alias
        if first_alias == ""
            my_columns = self.column_names - by
            other_columns = other.column_names - by
            intersection = my_columns & other_columns
            raise "Cannot merge because the following columns would overlap: #{intersection}" if intersection.size > 0
        else
            raise "The aliases can not have the same value"
        end
    end
    
    my_keys = {}
    self.each_with_index do |row, i|
        key = []
        by.each do |colname|
            key << row[colname]
        end
        
        my_keys[key] = i
    end
    
    merged_column_self  = (self.column_names - by)
    merged_column_other = (other.column_names - by)
    
    first_alias =  first_alias + "."     if first_alias.length > 0
    second_alias = second_alias + "."    if second_alias.length > 0
    
    merged_columns = merged_column_self.map { |colname| "#{first_alias}#{colname}" } + merged_column_other.map { |colname| "#{second_alias}#{colname}" }
    columns = by + merged_columns
    result = DataFrame.new(columns)
    other.each do |other_row|
        key = []
        by.each do |colname|
            key << other_row[colname]
        end
        
        my_row_index = my_keys[key]
        if my_row_index
            my_row = self.row(my_row_index)
            
            to_add = {}
            by.each do |colname|
                to_add[colname] = my_row[colname]
            end
            
            merged_column_self.each do |colname|
                to_add["#{first_alias}#{colname}"] = my_row[colname]
            end
            
            merged_column_other.each do |colname|
                to_add["#{second_alias}#{colname}"] = other_row[colname]
            end
            
            result << to_add
        end
    end
    
    return result
end

#rename_column!(old_name, new_name) ⇒ Object 



87
88
89
90
91
92
93
 
# File 'lib/rust/core/types/dataframe.rb', line 87

def rename_column!(old_name, new_name)
    raise "This DataFrame does not contain a column named #{old_name}" unless @labels.include?(old_name)
    raise "This DataFrame already contains a column named #{new_name}" if @labels.include?(new_name)
    
    @data[new_name.to_s] = @data.delete(old_name)
    @labels[@labels.index(old_name)] = new_name
end

#row(i) ⇒ Object 



37
38
39
40
41
42
43
 
# File 'lib/rust/core/types/dataframe.rb', line 37

def row(i)
    if i < 0 || i >= self.rows
        return nil
    else
        return @data.map { |label, values| [label, values[i]] }.to_h
    end
end

#rows ⇒ Object 



173
174
175
 
# File 'lib/rust/core/types/dataframe.rb', line 173

def rows
    @data.values[0].size
end

#select_columns(cols = nil) ⇒ Object Also known as: select_cols 



114
115
116
117
118
119
120
121
122
123
124
125
126
 
# File 'lib/rust/core/types/dataframe.rb', line 114

def select_columns(cols=nil)
    raise "You must specify either the columns you want to select or a selection block" if !cols && !block_given?
    
    result = self.clone
    @labels.each do |label|
        if cols
            result.delete_column(label) unless cols.include?(label)
        else
            result.delete_column(label) unless yield label
        end
    end
    return result
end

#select_rows ⇒ Object 



99
100
101
102
103
104
105
 
# File 'lib/rust/core/types/dataframe.rb', line 99

def select_rows
    result = DataFrame.new(self.column_names)
    self.each_with_index do |row, i|
        result << row if yield row, i
    end
    return result
end

#shuffle(*args) ⇒ Object 



53
54
55
56
57
58
59
60
61
62
63
64
65
 
# File 'lib/rust/core/types/dataframe.rb', line 53

def shuffle(*args)
    result = DataFrame.new(@labels)
    
    buffer = []
    self.each do |row|
        buffer << row
    end
    buffer.shuffle!(*args).each do |row|
        result << row
    end
    
    return result
end

#sort_by(column) ⇒ Object 



419
420
421
422
423
 
# File 'lib/rust/core/types/dataframe.rb', line 419

def sort_by(column)
    result = self.clone
    result.sort_by!(column)
    return result
end

#sort_by!(by) ⇒ Object 



425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
 
# File 'lib/rust/core/types/dataframe.rb', line 425

def sort_by!(by)
    copy = @data[by].clone
    copy.sort!
    
    indices = []
    @data[by].each_with_index do |value, i|
        index = copy.index(value)
        indices << index
        
        copy[index] = NilClass
    end
                
    (self.column_names - [by]).each do |column_name|
        sorted = []
        column = self.column(column_name)
        column_i = 0
        indices.each do |i|
            sorted[i] = column[column_i]
            column_i += 1
        end
        @data[column_name] = sorted
    end
    @data[by].sort!
end

#transform_column!(column) ⇒ Object 



95
96
97
 
# File 'lib/rust/core/types/dataframe.rb', line 95

def transform_column!(column)
    @data[column].map! { |e| yield e }
end

#uniq_by(by) ⇒ Object 



140
141
142
143
144
 
# File 'lib/rust/core/types/dataframe.rb', line 140

def uniq_by(by)
    result = self.clone
    result.uniq_by!(by)
    return result
end

#uniq_by!(by) ⇒ Object 



146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
 
# File 'lib/rust/core/types/dataframe.rb', line 146

def uniq_by!(by)
    my_keys = {}
    to_delete = []
    self.each_with_index do |row, i|
        key = []
        by.each do |colname|
            key << row[colname]
        end
        unless my_keys[key]
            my_keys[key] = i
        else
            to_delete << (i-to_delete.size)
        end
    end
    
    to_delete.each do |i|
        self.delete_row(i)
    end
    
    return self
end