red_amber 0.3.0 → 0.4.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.rubocop.yml +39 -20
- data/.yardopts +2 -0
- data/CHANGELOG.md +113 -0
- data/Gemfile +1 -1
- data/LICENSE +1 -1
- data/README.md +25 -26
- data/benchmark/basic.yml +2 -2
- data/benchmark/combine.yml +2 -2
- data/benchmark/dataframe.yml +2 -2
- data/benchmark/group.yml +2 -2
- data/benchmark/reshape.yml +2 -2
- data/benchmark/vector.yml +3 -0
- data/doc/DataFrame.md +32 -12
- data/doc/DataFrame_Comparison.md +65 -0
- data/doc/SubFrames.md +11 -0
- data/doc/Vector.md +207 -1
- data/doc/yard-templates/default/fulldoc/html/css/common.css +6 -0
- data/lib/red_amber/data_frame.rb +429 -75
- data/lib/red_amber/data_frame_combinable.rb +516 -66
- data/lib/red_amber/data_frame_displayable.rb +244 -14
- data/lib/red_amber/data_frame_indexable.rb +121 -18
- data/lib/red_amber/data_frame_loadsave.rb +78 -10
- data/lib/red_amber/data_frame_reshaping.rb +184 -14
- data/lib/red_amber/data_frame_selectable.rb +622 -66
- data/lib/red_amber/data_frame_variable_operation.rb +446 -34
- data/lib/red_amber/group.rb +187 -22
- data/lib/red_amber/helper.rb +70 -10
- data/lib/red_amber/refinements.rb +12 -5
- data/lib/red_amber/subframes.rb +1066 -0
- data/lib/red_amber/vector.rb +385 -11
- data/lib/red_amber/vector_aggregation.rb +312 -0
- data/lib/red_amber/vector_binary_element_wise.rb +387 -0
- data/lib/red_amber/vector_selectable.rb +217 -12
- data/lib/red_amber/vector_unary_element_wise.rb +436 -0
- data/lib/red_amber/vector_updatable.rb +278 -34
- data/lib/red_amber/version.rb +2 -1
- data/lib/red_amber.rb +13 -1
- data/red_amber.gemspec +2 -2
- metadata +13 -8
- data/doc/image/dataframe/reshaping_DataFrames.png +0 -0
- data/lib/red_amber/vector_functions.rb +0 -242
@@ -0,0 +1,1066 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
module RedAmber
|
4
|
+
# class SubFrames treats a set of subsets of a DataFrame
|
5
|
+
# [Experimental feature] Class SubFrames may be removed or be changed in the future.
|
6
|
+
class SubFrames
|
7
|
+
include Enumerable # may change to use Forwardable.
|
8
|
+
include Helper
|
9
|
+
|
10
|
+
using RefineArray
|
11
|
+
using RefineArrayLike
|
12
|
+
|
13
|
+
class << self
|
14
|
+
# Create SubFrames from a Group.
|
15
|
+
#
|
16
|
+
# [Experimental feature] this method may be removed or be changed in the future.
|
17
|
+
# @param group [Group]
|
18
|
+
# a Group to be used to create SubFrames.
|
19
|
+
# @return [SubFrames]
|
20
|
+
# a created SubFrames.
|
21
|
+
# @example
|
22
|
+
# dataframe
|
23
|
+
#
|
24
|
+
# # =>
|
25
|
+
# #<RedAmber::DataFrame : 6 x 3 Vectors, 0x000000000000fba4>
|
26
|
+
# x y z
|
27
|
+
# <uint8> <string> <boolean>
|
28
|
+
# 0 1 A false
|
29
|
+
# 1 2 A true
|
30
|
+
# 2 3 B false
|
31
|
+
# 3 4 B (nil)
|
32
|
+
# 4 5 B true
|
33
|
+
# 5 6 C false
|
34
|
+
#
|
35
|
+
# group = Group.new(dataframe, [:y])
|
36
|
+
# sf = SubFrames.by_group(group)
|
37
|
+
#
|
38
|
+
# # =>
|
39
|
+
# #<RedAmber::SubFrames : 0x000000000000fbb8>
|
40
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 3 Vectors, 0x000000000000fb7c>
|
41
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
42
|
+
# ---
|
43
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x000000000000fbcc>
|
44
|
+
# x y z
|
45
|
+
# <uint8> <string> <boolean>
|
46
|
+
# 0 1 A false
|
47
|
+
# 1 2 A true
|
48
|
+
# ---
|
49
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x000000000000fbe0>
|
50
|
+
# x y z
|
51
|
+
# <uint8> <string> <boolean>
|
52
|
+
# 0 3 B false
|
53
|
+
# 1 4 B (nil)
|
54
|
+
# 2 5 B true
|
55
|
+
# ---
|
56
|
+
# #<RedAmber::DataFrame : 1 x 3 Vectors, 0x000000000000fbf4>
|
57
|
+
# x y z
|
58
|
+
# <uint8> <string> <boolean>
|
59
|
+
# 0 6 C false
|
60
|
+
#
|
61
|
+
# @since 0.4.0
|
62
|
+
#
|
63
|
+
def by_group(group)
|
64
|
+
SubFrames.new(group.dataframe, group.filters)
|
65
|
+
end
|
66
|
+
|
67
|
+
# Create a new SubFrames object from a DataFrame and an array of indices.
|
68
|
+
#
|
69
|
+
# @api private
|
70
|
+
# @note this method doesn't check arguments.
|
71
|
+
# @param dataframe [DataFrame]
|
72
|
+
# a source dataframe.
|
73
|
+
# @param subset_indices [Array, Array<Vector>]
|
74
|
+
# an Array of numeric indices to create subsets of DataFrame.
|
75
|
+
# @return [SubFrames]
|
76
|
+
# a new SubFrames object.
|
77
|
+
# @since 0.4.0
|
78
|
+
#
|
79
|
+
def by_indices(dataframe, subset_indices)
|
80
|
+
instance = allocate
|
81
|
+
instance.instance_variable_set(:@baseframe, dataframe)
|
82
|
+
enum =
|
83
|
+
Enumerator.new(subset_indices.size) do |y|
|
84
|
+
subset_indices.each do |i|
|
85
|
+
y.yield dataframe.take(i)
|
86
|
+
end
|
87
|
+
end
|
88
|
+
instance.instance_variable_set(:@enum, enum)
|
89
|
+
instance
|
90
|
+
end
|
91
|
+
|
92
|
+
# Create a new SubFrames object from a DataFrame and an array of filters.
|
93
|
+
#
|
94
|
+
# @api private
|
95
|
+
# @note this method doesn't check arguments.
|
96
|
+
# @param dataframe [DataFrame]
|
97
|
+
# a source dataframe.
|
98
|
+
# @param subset_filters [Array, Array<Vector>]
|
99
|
+
# an Array of booleans to specify subsets of DataFrame.
|
100
|
+
# Each filters must have same length as dataframe.
|
101
|
+
# @return [SubFrames]
|
102
|
+
# a new SubFrames object.
|
103
|
+
# @since 0.4.0
|
104
|
+
#
|
105
|
+
def by_filters(dataframe, subset_filters)
|
106
|
+
instance = allocate
|
107
|
+
instance.instance_variable_set(:@baseframe, dataframe)
|
108
|
+
enum =
|
109
|
+
Enumerator.new(subset_filters.size) do |y|
|
110
|
+
subset_filters.each do |i|
|
111
|
+
y.yield dataframe.filter(i)
|
112
|
+
end
|
113
|
+
end
|
114
|
+
instance.instance_variable_set(:@enum, enum)
|
115
|
+
instance
|
116
|
+
end
|
117
|
+
|
118
|
+
# Create a new SubFrames from an Array of DataFrames.
|
119
|
+
#
|
120
|
+
# @api private
|
121
|
+
# @note dataframes must have same schema.
|
122
|
+
# @param dataframes [Array<DataFrame>]
|
123
|
+
# an array of DataFrames which have same schema.
|
124
|
+
# @return [SubFrames]
|
125
|
+
# a new SubFrames object.
|
126
|
+
# @since 0.4.0
|
127
|
+
#
|
128
|
+
def by_dataframes(dataframes)
|
129
|
+
instance = allocate
|
130
|
+
case Array(dataframes)
|
131
|
+
when [] || [nil]
|
132
|
+
instance.instance_variable_set(:@baseframe, DataFrame.new)
|
133
|
+
instance.instance_variable_set(:@frames, [])
|
134
|
+
enum = [].each
|
135
|
+
else
|
136
|
+
enum =
|
137
|
+
Enumerator.new(dataframes.size) do |y|
|
138
|
+
dataframes.each do |i|
|
139
|
+
y.yield i
|
140
|
+
end
|
141
|
+
end
|
142
|
+
instance.instance_variable_set(:@baseframe, enum.reduce(&:concatenate))
|
143
|
+
end
|
144
|
+
instance.instance_variable_set(:@enum, enum)
|
145
|
+
instance
|
146
|
+
end
|
147
|
+
|
148
|
+
private
|
149
|
+
|
150
|
+
# This method upgrades a iterating method from Enumerable to return SubFrames.
|
151
|
+
|
152
|
+
# @!macro [attach] define_subframable_method
|
153
|
+
#
|
154
|
+
# [Returns SubFrames] Use `#each.$1` if you want to get DataFrames by Array.
|
155
|
+
# Returns an Enumerator with no block given.
|
156
|
+
# @yieldparam dataframe [DataFrame]
|
157
|
+
# gives each element.
|
158
|
+
# @yieldreturn [Array<DataFrame>]
|
159
|
+
# the block should return DataFrames with same schema.
|
160
|
+
# @return [SubFrames]
|
161
|
+
# a new SubFrames.
|
162
|
+
#
|
163
|
+
def define_subframable_method(method)
|
164
|
+
define_method(method) do |&block|
|
165
|
+
return enum_for(:each) { size } unless block # rubocop:disable Lint/ToEnumArguments
|
166
|
+
|
167
|
+
self.class.by_dataframes(super(&block))
|
168
|
+
end
|
169
|
+
end
|
170
|
+
end
|
171
|
+
|
172
|
+
# Create a new SubFrames object from a DataFrame and an array of indices or filters.
|
173
|
+
#
|
174
|
+
# @overload initialize(dataframe, subset_specifier)
|
175
|
+
# Create a new SubFrames object.
|
176
|
+
#
|
177
|
+
# @param dataframe [DataFrame]
|
178
|
+
# a source dataframe.
|
179
|
+
# @param subset_specifier [Array<Vector>, Array<array-like>]
|
180
|
+
# an Array of numeric indices or boolean filters
|
181
|
+
# to create subsets of DataFrame.
|
182
|
+
# @return [SubFrames]
|
183
|
+
# new SubFrames.
|
184
|
+
# @example
|
185
|
+
# dataframe
|
186
|
+
#
|
187
|
+
# # =>
|
188
|
+
# #<RedAmber::DataFrame : 6 x 3 Vectors, 0x00000000000039e4>
|
189
|
+
# x y z
|
190
|
+
# <uint8> <string> <boolean>
|
191
|
+
# 0 1 A false
|
192
|
+
# 1 2 A true
|
193
|
+
# 2 3 B false
|
194
|
+
# 3 4 B (nil)
|
195
|
+
# 4 5 B true
|
196
|
+
# 5 6 C false
|
197
|
+
#
|
198
|
+
# SubFrames.new(dataframe, [[0, 2, 3], [4, 1]])
|
199
|
+
#
|
200
|
+
# # =>
|
201
|
+
# #<RedAmber::SubFrames : 0x0000000000003a34>
|
202
|
+
# @baseframe=#<RedAmber::DataFrame : 5 x 3 Vectors, 0x0000000000003a48>
|
203
|
+
# 2 SubFrames: [3, 2] in sizes.
|
204
|
+
# ---
|
205
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x0000000000003a5c>
|
206
|
+
# x y z
|
207
|
+
# <uint8> <string> <boolean>
|
208
|
+
# 0 1 A false
|
209
|
+
# 1 3 B false
|
210
|
+
# 2 4 B (nil)
|
211
|
+
# ---
|
212
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x0000000000003a70>
|
213
|
+
# x y z
|
214
|
+
# <uint8> <string> <boolean>
|
215
|
+
# 0 5 B true
|
216
|
+
# 1 2 A true
|
217
|
+
#
|
218
|
+
# @overload initialize(dataframe)
|
219
|
+
# Create a new SubFrames object by block.
|
220
|
+
#
|
221
|
+
# @param dataframe [DataFrame]
|
222
|
+
# a source dataframe.
|
223
|
+
# @yieldparam dataframe [DataFrame]
|
224
|
+
# the block is called with `dataframe`.
|
225
|
+
# @yieldreturn [Array<numeric_array_like>, Array<boolean_array_like>]
|
226
|
+
# an Array of index or boolean array-likes to create subsets of DataFrame.
|
227
|
+
# All array-likes are responsible to #numeric? or #boolean?.
|
228
|
+
# @return [SubFrames]
|
229
|
+
# a new SubFrames object.
|
230
|
+
# @example
|
231
|
+
# SubFrames.new(dataframe) do |df|
|
232
|
+
# booleans = df[:z]
|
233
|
+
# [booleans, !booleans]
|
234
|
+
# end
|
235
|
+
#
|
236
|
+
# # =>
|
237
|
+
# #<RedAmber::SubFrames : 0x0000000000003aac>
|
238
|
+
# @baseframe=#<RedAmber::DataFrame : 5 x 3 Vectors, 0x0000000000003ac0>
|
239
|
+
# 2 SubFrames: [2, 3] in sizes.
|
240
|
+
# ---
|
241
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x0000000000003ad4>
|
242
|
+
# x y z
|
243
|
+
# <uint8> <string> <boolean>
|
244
|
+
# 0 2 A true
|
245
|
+
# 1 5 B true
|
246
|
+
# ---
|
247
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x0000000000003ae8>
|
248
|
+
# x y z
|
249
|
+
# <uint8> <string> <boolean>
|
250
|
+
# 0 1 A false
|
251
|
+
# 1 3 B false
|
252
|
+
# 2 6 C false
|
253
|
+
#
|
254
|
+
# @since 0.4.0
|
255
|
+
#
|
256
|
+
def initialize(dataframe, subset_specifier = nil, &block)
|
257
|
+
unless dataframe.is_a?(DataFrame)
|
258
|
+
raise SubFramesArgumentError, "not a DataFrame: #{dataframe}"
|
259
|
+
end
|
260
|
+
|
261
|
+
if block
|
262
|
+
unless subset_specifier.nil?
|
263
|
+
raise SubFramesArgumentError, 'Must not specify both arguments and block.'
|
264
|
+
end
|
265
|
+
|
266
|
+
subset_specifier = yield(dataframe)
|
267
|
+
end
|
268
|
+
|
269
|
+
if dataframe.empty? || subset_specifier.nil? || subset_specifier.empty?
|
270
|
+
@baseframe = DataFrame.new
|
271
|
+
@frames = []
|
272
|
+
@enum = @frames.each
|
273
|
+
else
|
274
|
+
@baseframe = nil
|
275
|
+
@enum =
|
276
|
+
Enumerator.new(subset_specifier.size) do |yielder|
|
277
|
+
subset_specifier.map do |i|
|
278
|
+
df =
|
279
|
+
if i.numeric?
|
280
|
+
dataframe.take(i)
|
281
|
+
elsif i.boolean?
|
282
|
+
dataframe.filter(i)
|
283
|
+
else
|
284
|
+
raise SubFramesArgumentError, "illegal type: #{i}"
|
285
|
+
end
|
286
|
+
yielder.yield df
|
287
|
+
end
|
288
|
+
end
|
289
|
+
end
|
290
|
+
end
|
291
|
+
|
292
|
+
# Return concatenated SubFrames as a DataDrame.
|
293
|
+
#
|
294
|
+
# Once evaluated, memorize it as @baseframe.
|
295
|
+
# @return [DataFrame]
|
296
|
+
# a concatenated DataFrame.
|
297
|
+
# @since 0.4.0
|
298
|
+
#
|
299
|
+
def baseframe
|
300
|
+
@baseframe ||= reduce(&:concatenate)
|
301
|
+
end
|
302
|
+
alias_method :concatenate, :baseframe
|
303
|
+
alias_method :concat, :baseframe
|
304
|
+
|
305
|
+
# Iterates over sub DataFrames or returns an Enumerator.
|
306
|
+
#
|
307
|
+
# This method will memorize sub DataFrames and always returns the same object.
|
308
|
+
# The Class SubFrames is including Enumerable module.
|
309
|
+
# So many methods in Enumerable are available.
|
310
|
+
#
|
311
|
+
# @overload each
|
312
|
+
# Returns a new Enumerator if no block given.
|
313
|
+
#
|
314
|
+
# @return [Enumerator]
|
315
|
+
# Enumerator of each elements.
|
316
|
+
#
|
317
|
+
# @overload each
|
318
|
+
# When a block given, passes each sub DataFrames to the block.
|
319
|
+
#
|
320
|
+
# @yieldparam subframe [DataFrame]
|
321
|
+
# passes sub DataFrame by a block parameter.
|
322
|
+
# @yieldreturn [Object]
|
323
|
+
# evaluated result value from the block.
|
324
|
+
# @return [self]
|
325
|
+
# returns self.
|
326
|
+
#
|
327
|
+
# @example Returns Enumerator
|
328
|
+
# sf.each
|
329
|
+
#
|
330
|
+
# # =>
|
331
|
+
# #<Enumerator: ...>
|
332
|
+
#
|
333
|
+
# @example `to_a` from Enumerable.
|
334
|
+
# sf.to_a
|
335
|
+
#
|
336
|
+
# # =>
|
337
|
+
# [#<RedAmber::DataFrame : 2 x 3 Vectors, 0x000000000002a120>
|
338
|
+
# x y z
|
339
|
+
# <uint8> <string> <boolean>
|
340
|
+
# 0 1 A false
|
341
|
+
# 1 2 A true
|
342
|
+
# ,
|
343
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x000000000002a134>
|
344
|
+
# x y z
|
345
|
+
# <uint8> <string> <boolean>
|
346
|
+
# 0 3 B false
|
347
|
+
# 1 4 B (nil)
|
348
|
+
# 2 5 B true
|
349
|
+
# ,
|
350
|
+
# #<RedAmber::DataFrame : 1 x 3 Vectors, 0x000000000002a148>
|
351
|
+
# x y z
|
352
|
+
# <uint8> <string> <boolean>
|
353
|
+
# 0 6 C false
|
354
|
+
# ]
|
355
|
+
#
|
356
|
+
# @example Concatenate SubFrames. This example is used in #concatenate.
|
357
|
+
# sf.reduce(&:concatenate)
|
358
|
+
#
|
359
|
+
# # =>
|
360
|
+
# #<RedAmber::DataFrame : 6 x 3 Vectors, 0x000000000004883c>
|
361
|
+
# x y z
|
362
|
+
# <uint8> <string> <boolean>
|
363
|
+
# 0 1 A false
|
364
|
+
# 1 2 A true
|
365
|
+
# 2 3 B false
|
366
|
+
# 3 4 B (nil)
|
367
|
+
# 4 5 B true
|
368
|
+
# 5 6 C false
|
369
|
+
#
|
370
|
+
# @since 0.4.0
|
371
|
+
#
|
372
|
+
def each(&block)
|
373
|
+
return enum_for(__method__) { size } unless block
|
374
|
+
|
375
|
+
frames.each(&block)
|
376
|
+
nil
|
377
|
+
end
|
378
|
+
|
379
|
+
# Aggregate SubFrames to create a DataFrame.
|
380
|
+
#
|
381
|
+
# This method will check if built-in aggregation function is used.
|
382
|
+
# @todo Support user-defined aggregation functions.
|
383
|
+
#
|
384
|
+
# @overload aggregate(group_keys, aggregations)
|
385
|
+
#
|
386
|
+
# Aggregate SubFrames for first values of the columns of
|
387
|
+
# `group_keys` and the aggregated results of key-function pairs.
|
388
|
+
#
|
389
|
+
# @param group_keys [Symbol, String, Array<Symbol, String>]
|
390
|
+
# group key name(s) to output values.
|
391
|
+
# @param aggregations [Hash<Array<Symbol, String> => Array<:Symbol>>]
|
392
|
+
# a Hash of variable (column) name and
|
393
|
+
# Vector aggregate function name to apply.
|
394
|
+
# @return [DataFrame]
|
395
|
+
# an aggregated DataFrame.
|
396
|
+
# @example
|
397
|
+
# subframes
|
398
|
+
#
|
399
|
+
# # =>
|
400
|
+
# #<RedAmber::SubFrames : 0x0000000000003980>
|
401
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 3 Vectors, 0x0000000000003994>
|
402
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
403
|
+
# ---
|
404
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x00000000000039a8>
|
405
|
+
# x y z
|
406
|
+
# <uint8> <string> <boolean>
|
407
|
+
# 0 1 A false
|
408
|
+
# 1 2 A true
|
409
|
+
# ---
|
410
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x00000000000039bc>
|
411
|
+
# x y z
|
412
|
+
# <uint8> <string> <boolean>
|
413
|
+
# 0 3 B false
|
414
|
+
# 1 4 B (nil)
|
415
|
+
# 2 5 B true
|
416
|
+
# ---
|
417
|
+
# #<RedAmber::DataFrame : 1 x 3 Vectors, 0x00000000000039d0>
|
418
|
+
# x y z
|
419
|
+
# <uint8> <string> <boolean>
|
420
|
+
# 0 6 C false
|
421
|
+
#
|
422
|
+
# subframes.aggregate(:y, { x: :sum })
|
423
|
+
#
|
424
|
+
# # =>
|
425
|
+
# #<RedAmber::DataFrame : 3 x 2 Vectors, 0x0000000000003b24>
|
426
|
+
# y sum_x
|
427
|
+
# <string> <uint8>
|
428
|
+
# 0 A 3
|
429
|
+
# 1 B 12
|
430
|
+
# 2 C 6
|
431
|
+
#
|
432
|
+
# @overload aggregate(group_keys, aggregations)
|
433
|
+
#
|
434
|
+
# Aggregate SubFrames for first values of the columns of
|
435
|
+
# `group_keys` and the aggregated results of all combinations
|
436
|
+
# of supplied keys and functions.
|
437
|
+
#
|
438
|
+
# @param group_keys [Symbol, String, Array<Symbol, String>]
|
439
|
+
# group key name(s) to output values.
|
440
|
+
# @param aggregations [Array[Array<Symbol, String>, Array<:Symbol>]]
|
441
|
+
# an Array of Array of variable (column) names and
|
442
|
+
# Array of Vector aggregate function names to apply.
|
443
|
+
# @return [DataFrame]
|
444
|
+
# an aggregated DataFrame.
|
445
|
+
# @example
|
446
|
+
# sf.aggregate(:y, [[:x, :z], [:count, :sum]])
|
447
|
+
#
|
448
|
+
# # =>
|
449
|
+
# #<RedAmber::DataFrame : 3 x 5 Vectors, 0x000000000000fcbc>
|
450
|
+
# y count_x count_z sum_x sum_z
|
451
|
+
# <string> <uint8> <uint8> <uint8> <uint8>
|
452
|
+
# 0 A 2 2 3 1
|
453
|
+
# 1 B 3 2 12 1
|
454
|
+
# 2 C 1 1 6 0
|
455
|
+
#
|
456
|
+
# @since 0.4.0
|
457
|
+
#
|
458
|
+
def aggregate(group_keys, aggregations)
|
459
|
+
aggregator =
|
460
|
+
case aggregations
|
461
|
+
in Hash
|
462
|
+
sf = self
|
463
|
+
aggregations.map do |key, func|
|
464
|
+
unless Vector.aggregate?(func)
|
465
|
+
raise SubFramesArgumentError, "not an aggregation function: #{func}"
|
466
|
+
end
|
467
|
+
|
468
|
+
["#{func}_#{key}", sf.each.map { |df| df[key].send(func) }]
|
469
|
+
end
|
470
|
+
in [Array => keys, Array => functions]
|
471
|
+
functions.each do |func|
|
472
|
+
unless Vector.aggregate?(func)
|
473
|
+
raise SubFramesArgumentError, "not an aggregation function: #{func}"
|
474
|
+
end
|
475
|
+
end
|
476
|
+
sf = self
|
477
|
+
functions.product(keys).map do |func, key|
|
478
|
+
["#{func}_#{key}", sf.each.map { |df| df[key].send(func) }]
|
479
|
+
end
|
480
|
+
else
|
481
|
+
raise SubFramesArgumentError, "invalid argument: #{aggregations}"
|
482
|
+
end
|
483
|
+
|
484
|
+
if group_keys.empty?
|
485
|
+
DataFrame.new(aggregator)
|
486
|
+
else
|
487
|
+
baseframe
|
488
|
+
.pick(group_keys)
|
489
|
+
.slice(offset_indices)
|
490
|
+
.assign(aggregator)
|
491
|
+
end
|
492
|
+
end
|
493
|
+
|
494
|
+
# Returns a SubFrames containing DataFrames returned by the block.
|
495
|
+
#
|
496
|
+
# @example Map as it is.
|
497
|
+
# subframes
|
498
|
+
#
|
499
|
+
# # =>
|
500
|
+
# #<RedAmber::SubFrames : 0x000000000001359c>
|
501
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 3 Vectors, 0x00000000000135b0>
|
502
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
503
|
+
# ---
|
504
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x00000000000135c4>
|
505
|
+
# x y z
|
506
|
+
# <uint8> <string> <boolean>
|
507
|
+
# 0 1 A false
|
508
|
+
# 1 2 A true
|
509
|
+
# ---
|
510
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x00000000000135d8>
|
511
|
+
# x y z
|
512
|
+
# <uint8> <string> <boolean>
|
513
|
+
# 0 3 B false
|
514
|
+
# 1 4 B (nil)
|
515
|
+
# 2 5 B true
|
516
|
+
# ---
|
517
|
+
# #<RedAmber::DataFrame : 1 x 3 Vectors, 0x00000000000135ec>
|
518
|
+
# x y z
|
519
|
+
# <uint8> <string> <boolean>
|
520
|
+
# 0 6 C false
|
521
|
+
#
|
522
|
+
# subframes.map { _1 }
|
523
|
+
#
|
524
|
+
# # This will create a new SubFrame and a new baseframe,
|
525
|
+
# # But each element DataFrames are re-used.
|
526
|
+
# # =>
|
527
|
+
# #<RedAmber::SubFrames : 0x000000000001e6cc>
|
528
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 3 Vectors, 0x000000000001e6e0>
|
529
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
530
|
+
# ---
|
531
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x00000000000135c4>
|
532
|
+
# x y z
|
533
|
+
# <uint8> <string> <boolean>
|
534
|
+
# 0 1 A false
|
535
|
+
# 1 2 A true
|
536
|
+
# ---
|
537
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x00000000000135d8>
|
538
|
+
# x y z
|
539
|
+
# <uint8> <string> <boolean>
|
540
|
+
# 0 3 B false
|
541
|
+
# 1 4 B (nil)
|
542
|
+
# 2 5 B true
|
543
|
+
# ---
|
544
|
+
# #<RedAmber::DataFrame : 1 x 3 Vectors, 0x00000000000135ec>
|
545
|
+
# x y z
|
546
|
+
# <uint8> <string> <boolean>
|
547
|
+
# 0 6 C false
|
548
|
+
#
|
549
|
+
# @example Assign a new column.
|
550
|
+
# subframes.map { |df| df.assign(x_plus1: df[:x] + 1) }
|
551
|
+
#
|
552
|
+
# # =>
|
553
|
+
# #<RedAmber::SubFrames : 0x0000000000040948>
|
554
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 4 Vectors, 0x000000000004095c>
|
555
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
556
|
+
# ---
|
557
|
+
# #<RedAmber::DataFrame : 2 x 4 Vectors, 0x0000000000040970>
|
558
|
+
# x y z x_plus1
|
559
|
+
# <uint8> <string> <boolean> <uint8>
|
560
|
+
# 0 1 A false 2
|
561
|
+
# 1 2 A true 3
|
562
|
+
# ---
|
563
|
+
# #<RedAmber::DataFrame : 3 x 4 Vectors, 0x0000000000040984>
|
564
|
+
# x y z x_plus1
|
565
|
+
# <uint8> <string> <boolean> <uint8>
|
566
|
+
# 0 3 B false 4
|
567
|
+
# 1 4 B (nil) 5
|
568
|
+
# 2 5 B true 6
|
569
|
+
# ---
|
570
|
+
# #<RedAmber::DataFrame : 1 x 4 Vectors, 0x0000000000040998>
|
571
|
+
# x y z x_plus1
|
572
|
+
# <uint8> <string> <boolean> <uint8>
|
573
|
+
# 0 6 C false 7
|
574
|
+
#
|
575
|
+
# @since 0.4.0
|
576
|
+
#
|
577
|
+
define_subframable_method :map
|
578
|
+
alias_method :collect, :map
|
579
|
+
|
580
|
+
# Update existing column(s) or create new columns(s) for each DataFrames in self.
|
581
|
+
#
|
582
|
+
# Column values are updated by an oveloaded common operation.
|
583
|
+
#
|
584
|
+
# @overload assign(key)
|
585
|
+
# Assign a column by argument and block.
|
586
|
+
#
|
587
|
+
# @param key [Symbol, String]
|
588
|
+
# a key of column to assign.
|
589
|
+
# @yieldparam dataframe [DataFrame]
|
590
|
+
# gives overloaded dataframe in self to the block.
|
591
|
+
# @yieldreturn [Vector, Array, Arrow::Array]
|
592
|
+
# an updated column value which are overloaded.
|
593
|
+
# @return [SubFrames]
|
594
|
+
# a new SubFrames object with updated DataFrames.
|
595
|
+
# @example
|
596
|
+
# subframes
|
597
|
+
#
|
598
|
+
# # =>
|
599
|
+
# #<RedAmber::SubFrames : 0x000000000000c33c>
|
600
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 3 Vectors, 0x000000000000c350>
|
601
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
602
|
+
# ---
|
603
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x000000000000c364>
|
604
|
+
# x y z
|
605
|
+
# <uint8> <string> <boolean>
|
606
|
+
# 0 1 A false
|
607
|
+
# 1 2 A true
|
608
|
+
# ---
|
609
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x000000000000c378>
|
610
|
+
# x y z
|
611
|
+
# <uint8> <string> <boolean>
|
612
|
+
# 0 3 B false
|
613
|
+
# 1 4 B (nil)
|
614
|
+
# 2 5 B true
|
615
|
+
# ---
|
616
|
+
# #<RedAmber::DataFrame : 1 x 3 Vectors, 0x000000000000c38c>
|
617
|
+
# x y z
|
618
|
+
# <uint8> <string> <boolean>
|
619
|
+
# 0 6 C false
|
620
|
+
#
|
621
|
+
# subframes.assign(:x_plus1) { x + 1 }
|
622
|
+
#
|
623
|
+
# # =>
|
624
|
+
# #<RedAmber::SubFrames : 0x000000000000c3a0>
|
625
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 4 Vectors, 0x000000000000c3b4>
|
626
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
627
|
+
# ---
|
628
|
+
# #<RedAmber::DataFrame : 2 x 4 Vectors, 0x000000000000c3c8>
|
629
|
+
# x y z x_plus1
|
630
|
+
# <uint8> <string> <boolean> <uint8>
|
631
|
+
# 0 1 A false 2
|
632
|
+
# 1 2 A true 3
|
633
|
+
# ---
|
634
|
+
# #<RedAmber::DataFrame : 3 x 4 Vectors, 0x000000000000c3dc>
|
635
|
+
# x y z x_plus1
|
636
|
+
# <uint8> <string> <boolean> <uint8>
|
637
|
+
# 0 3 B false 4
|
638
|
+
# 1 4 B (nil) 5
|
639
|
+
# 2 5 B true 6
|
640
|
+
# ---
|
641
|
+
# #<RedAmber::DataFrame : 1 x 4 Vectors, 0x000000000000c3f0>
|
642
|
+
# x y z x_plus1
|
643
|
+
# <uint8> <string> <boolean> <uint8>
|
644
|
+
# 0 6 C false 7
|
645
|
+
#
|
646
|
+
# @overload assign(keys)
|
647
|
+
# Assign columns by arguments and block.
|
648
|
+
#
|
649
|
+
# @param keys [Array<Symbol, String>]
|
650
|
+
# keys of columns to assign.
|
651
|
+
# @yieldparam dataframe [DataFrame]
|
652
|
+
# gives overloaded dataframes in self to the block.
|
653
|
+
# @yieldreturn [Array<Vector, Array, Arrow::Array>]
|
654
|
+
# an updated column values which are overloaded.
|
655
|
+
# @return [SubFrames]
|
656
|
+
# a new SubFrames object with updated DataFrames.
|
657
|
+
# @example
|
658
|
+
# subframes.assign(:sum_x, :frac_x) do
|
659
|
+
# group_sum = x.sum
|
660
|
+
# [[group_sum] * size, x / s.to_f]
|
661
|
+
# end
|
662
|
+
#
|
663
|
+
# # =>
|
664
|
+
# #<RedAmber::SubFrames : 0x000000000000fce4>
|
665
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 5 Vectors, 0x000000000000fcf8>
|
666
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
667
|
+
# ---
|
668
|
+
# #<RedAmber::DataFrame : 2 x 5 Vectors, 0x000000000000fd0c>
|
669
|
+
# x y z sum_x frac_x
|
670
|
+
# <uint8> <string> <boolean> <uint8> <double>
|
671
|
+
# 0 1 A false 3 0.33
|
672
|
+
# 1 2 A true 3 0.67
|
673
|
+
# ---
|
674
|
+
# #<RedAmber::DataFrame : 3 x 5 Vectors, 0x000000000000fd20>
|
675
|
+
# x y z sum_x frac_x
|
676
|
+
# <uint8> <string> <boolean> <uint8> <double>
|
677
|
+
# 0 3 B false 12 0.25
|
678
|
+
# 1 4 B (nil) 12 0.33
|
679
|
+
# 2 5 B true 12 0.42
|
680
|
+
# ---
|
681
|
+
# #<RedAmber::DataFrame : 1 x 5 Vectors, 0x000000000000fd34>
|
682
|
+
# x y z sum_x frac_x
|
683
|
+
# <uint8> <string> <boolean> <uint8> <double>
|
684
|
+
# 0 6 C false 6 1.0
|
685
|
+
#
|
686
|
+
# @overload assign
|
687
|
+
# Assign column(s) by block.
|
688
|
+
#
|
689
|
+
# @yieldparam dataframe [DataFrame]
|
690
|
+
# gives overloaded dataframes in self to the block.
|
691
|
+
# @yieldreturn [Hash, Array]
|
692
|
+
# pairs of keys and column values which are overloaded.
|
693
|
+
# @return [SubFrames]
|
694
|
+
# a new SubFrames object with updated DataFrames.
|
695
|
+
# @example Compute 'x * z' when (true, not_true) = (1, 0) in z
|
696
|
+
# subframes.assign do
|
697
|
+
# { 'x*z': x * z.if_else(1, 0) }
|
698
|
+
# end
|
699
|
+
#
|
700
|
+
# # =>
|
701
|
+
# #<RedAmber::SubFrames : 0x000000000000fd98>
|
702
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 4 Vectors, 0x000000000000fdac>
|
703
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
704
|
+
# ---
|
705
|
+
# #<RedAmber::DataFrame : 2 x 4 Vectors, 0x000000000000fdc0>
|
706
|
+
# x y z x*z
|
707
|
+
# <uint8> <string> <boolean> <uint8>
|
708
|
+
# 0 1 A false 0
|
709
|
+
# 1 2 A true 2
|
710
|
+
# ---
|
711
|
+
# #<RedAmber::DataFrame : 3 x 4 Vectors, 0x000000000000fdd4>
|
712
|
+
# x y z x*z
|
713
|
+
# <uint8> <string> <boolean> <uint8>
|
714
|
+
# 0 3 B false 0
|
715
|
+
# 1 4 B (nil) (nil)
|
716
|
+
# 2 5 B true 5
|
717
|
+
# ---
|
718
|
+
# #<RedAmber::DataFrame : 1 x 4 Vectors, 0x000000000000fde8>
|
719
|
+
# x y z x*z
|
720
|
+
# <uint8> <string> <boolean> <uint8>
|
721
|
+
# 0 6 C false 0
|
722
|
+
#
|
723
|
+
# @since 0.4.0
|
724
|
+
#
|
725
|
+
def assign(...)
|
726
|
+
map { |df| df.assign(...) }
|
727
|
+
end
|
728
|
+
|
729
|
+
# Returns a SubFrames containing DataFrames selected by the block.
|
730
|
+
#
|
731
|
+
# With a block given, calls the block with successive DataFrames;
|
732
|
+
# returns a SubFrames of those DataFrames for
|
733
|
+
# which the block returns a truthy value.
|
734
|
+
#
|
735
|
+
# @example Select all.
|
736
|
+
# subframes.select { true }
|
737
|
+
#
|
738
|
+
# # =>
|
739
|
+
# #<RedAmber::SubFrames : 0x0000000000003a84>
|
740
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 3 Vectors, 0x0000000000003a98>
|
741
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
742
|
+
# ---
|
743
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x0000000000003a0c>
|
744
|
+
# x y z
|
745
|
+
# <uint8> <string> <boolean>
|
746
|
+
# 0 1 A false
|
747
|
+
# 1 2 A true
|
748
|
+
# ---
|
749
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x0000000000003a20>
|
750
|
+
# x y z
|
751
|
+
# <uint8> <string> <boolean>
|
752
|
+
# 0 3 B false
|
753
|
+
# 1 4 B (nil)
|
754
|
+
# 2 5 B true
|
755
|
+
# ---
|
756
|
+
# #<RedAmber::DataFrame : 1 x 3 Vectors, 0x0000000000003a34>
|
757
|
+
# x y z
|
758
|
+
# <uint8> <string> <boolean>
|
759
|
+
# 0 6 C false
|
760
|
+
#
|
761
|
+
# @example Select nothing.
|
762
|
+
# subframes.select { false }
|
763
|
+
#
|
764
|
+
# # =>
|
765
|
+
# #<RedAmber::SubFrames : 0x00000000000238c0>
|
766
|
+
# @baseframe=#<RedAmber::DataFrame : (empty), 0x00000000000238d4>
|
767
|
+
# 0 SubFrame: [] in size.
|
768
|
+
# ---
|
769
|
+
#
|
770
|
+
# @example Select if Vector `:z` has any true.
|
771
|
+
# subframes.select { |df| df[:z].any? }
|
772
|
+
#
|
773
|
+
# # =>
|
774
|
+
# #<RedAmber::SubFrames : 0x000000000000fba4>
|
775
|
+
# @baseframe=#<RedAmber::DataFrame : 3 x 3 Vectors, 0x000000000000fbb8>
|
776
|
+
# 2 SubFrames: [2, 1] in sizes.
|
777
|
+
# ---
|
778
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x0000000000003a0c>
|
779
|
+
# x y z
|
780
|
+
# <uint8> <string> <boolean>
|
781
|
+
# 0 1 A false
|
782
|
+
# 1 2 A true
|
783
|
+
# ---
|
784
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x0000000000003a20>
|
785
|
+
# x y z
|
786
|
+
# <uint8> <string> <boolean>
|
787
|
+
# 0 3 B false
|
788
|
+
# 1 4 B (nil)
|
789
|
+
# 2 5 B true
|
790
|
+
#
|
791
|
+
# @since 0.4.0
|
792
|
+
#
|
793
|
+
define_subframable_method :select
|
794
|
+
alias_method :filter, :select
|
795
|
+
alias_method :find_all, :select
|
796
|
+
|
797
|
+
# Returns a SubFrames containing DataFrames rejected by the block.
|
798
|
+
#
|
799
|
+
# With a block given, calls the block with successive DataFrames;
|
800
|
+
# returns a SubFrames of those DataFrames for
|
801
|
+
# which the block returns nil or false.
|
802
|
+
# @example Reject all.
|
803
|
+
# subframes.reject { true }
|
804
|
+
#
|
805
|
+
# # =>
|
806
|
+
# #<RedAmber::SubFrames : 0x00000000000238c0>
|
807
|
+
# @baseframe=#<RedAmber::DataFrame : (empty), 0x00000000000238d4>
|
808
|
+
# 0 SubFrame: [] in size.
|
809
|
+
# ---
|
810
|
+
#
|
811
|
+
# @example Reject nothing.
|
812
|
+
# subframes.reject { false }
|
813
|
+
#
|
814
|
+
# # =>
|
815
|
+
# #<RedAmber::SubFrames : 0x0000000000003a84>
|
816
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 3 Vectors, 0x0000000000003a98>
|
817
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
818
|
+
# ---
|
819
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x0000000000003a0c>
|
820
|
+
# x y z
|
821
|
+
# <uint8> <string> <boolean>
|
822
|
+
# 0 1 A false
|
823
|
+
# 1 2 A true
|
824
|
+
# ---
|
825
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x0000000000003a20>
|
826
|
+
# x y z
|
827
|
+
# <uint8> <string> <boolean>
|
828
|
+
# 0 3 B false
|
829
|
+
# 1 4 B (nil)
|
830
|
+
# 2 5 B true
|
831
|
+
# ---
|
832
|
+
# #<RedAmber::DataFrame : 1 x 3 Vectors, 0x0000000000003a34>
|
833
|
+
# x y z
|
834
|
+
# <uint8> <string> <boolean>
|
835
|
+
# 0 6 C false
|
836
|
+
#
|
837
|
+
# @example Reject if Vector `:z` has any true.
|
838
|
+
# subframes.reject { |df| df[:z].any? }
|
839
|
+
#
|
840
|
+
# # =>
|
841
|
+
# #<RedAmber::SubFrames : 0x0000000000038d74>
|
842
|
+
# @baseframe=#<RedAmber::DataFrame : 1 x 3 Vectors, 0x000000000001ad10>
|
843
|
+
# 1 SubFrame: [1] in size.
|
844
|
+
# ---
|
845
|
+
# #<RedAmber::DataFrame : 1 x 3 Vectors, 0x000000000001ad10>
|
846
|
+
# x y z
|
847
|
+
# <uint8> <string> <boolean>
|
848
|
+
# 0 6 C false
|
849
|
+
#
|
850
|
+
# @since 0.4.0
|
851
|
+
#
|
852
|
+
define_subframable_method :reject
|
853
|
+
|
854
|
+
# Returns a SubFrames containing truthy DataFrames returned by the block.
|
855
|
+
#
|
856
|
+
# With a block given, calls the block with successive DataFrames;
|
857
|
+
# returns a SubFrames of those DataFrames for
|
858
|
+
# which the block returns nil or false.
|
859
|
+
# @example Filter for size is larger than 1 and append number to column 'y'.
|
860
|
+
# subframes.filter_map do |df|
|
861
|
+
# if df.size > 1
|
862
|
+
# df.assign(:y) do
|
863
|
+
# y.merge(indices('1'), sep: '')
|
864
|
+
# end
|
865
|
+
# end
|
866
|
+
# end
|
867
|
+
#
|
868
|
+
# # =>
|
869
|
+
# #<RedAmber::SubFrames : 0x000000000001da88>
|
870
|
+
# @baseframe=#<RedAmber::DataFrame : 5 x 3 Vectors, 0x000000000001da9c>
|
871
|
+
# 2 SubFrames: [2, 3] in sizes.
|
872
|
+
# ---
|
873
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x000000000001dab0>
|
874
|
+
# x y z
|
875
|
+
# <uint8> <string> <boolean>
|
876
|
+
# 0 1 A1 false
|
877
|
+
# 1 2 A2 true
|
878
|
+
# ---
|
879
|
+
# #<RedAmber::DataFrame : 3 x 3 Vectors, 0x000000000001dac4>
|
880
|
+
# x y z
|
881
|
+
# <uint8> <string> <boolean>
|
882
|
+
# 0 3 B1 false
|
883
|
+
# 1 4 B2 (nil)
|
884
|
+
# 2 5 B3 true
|
885
|
+
#
|
886
|
+
# @since 0.4.0
|
887
|
+
#
|
888
|
+
define_subframable_method :filter_map
|
889
|
+
|
890
|
+
# Number of subsets.
|
891
|
+
#
|
892
|
+
# @return [Integer]
|
893
|
+
# number of subsets in self.
|
894
|
+
# @since 0.4.0
|
895
|
+
#
|
896
|
+
def size
|
897
|
+
@size ||= @enum.size
|
898
|
+
end
|
899
|
+
|
900
|
+
# Size list of subsets.
|
901
|
+
#
|
902
|
+
# @return [Array<Integer>]
|
903
|
+
# sizes of sub DataFrames.
|
904
|
+
# @since 0.4.0
|
905
|
+
#
|
906
|
+
def sizes
|
907
|
+
@sizes ||= @enum.map(&:size)
|
908
|
+
end
|
909
|
+
|
910
|
+
# Indices at the top of each sub DataFrames.
|
911
|
+
#
|
912
|
+
# @return [Array<Integer>]
|
913
|
+
# indices of offset of each sub DataFrames.
|
914
|
+
# @example When `sizes` is [2, 3, 1].
|
915
|
+
# sf.offset_indices # => [0, 2, 5]
|
916
|
+
# @since 0.4.0
|
917
|
+
#
|
918
|
+
def offset_indices
|
919
|
+
sum = 0
|
920
|
+
sizes.map do |size|
|
921
|
+
sum += size
|
922
|
+
sum - size
|
923
|
+
end
|
924
|
+
end
|
925
|
+
|
926
|
+
# Test if subset is empty?.
|
927
|
+
#
|
928
|
+
# @return [true, false]
|
929
|
+
# true if self is an empty subset.
|
930
|
+
# @since 0.4.0
|
931
|
+
#
|
932
|
+
def empty?
|
933
|
+
size.zero?
|
934
|
+
end
|
935
|
+
|
936
|
+
# Test if self has only one subset and it is comprehensive.
|
937
|
+
#
|
938
|
+
# @return [true, false]
|
939
|
+
# true if only member of self is equal to universal DataFrame.
|
940
|
+
# @since 0.4.0
|
941
|
+
#
|
942
|
+
def universal?
|
943
|
+
size == 1 && @enum.first == baseframe
|
944
|
+
end
|
945
|
+
|
946
|
+
# Return string representation of self.
|
947
|
+
#
|
948
|
+
# @param limit [Integer]
|
949
|
+
# maximum number of DataFrames to show.
|
950
|
+
# @return [String]
|
951
|
+
# return string representation of each sub DataFrame.
|
952
|
+
# @example
|
953
|
+
# df
|
954
|
+
#
|
955
|
+
# # =>
|
956
|
+
# #<RedAmber::DataFrame : 6 x 3 Vectors, 0x000000000000caa8>
|
957
|
+
# x y z
|
958
|
+
# <uint8> <string> <boolean>
|
959
|
+
# 0 1 A false
|
960
|
+
# 1 2 A true
|
961
|
+
# 2 3 B false
|
962
|
+
# 3 4 B (nil)
|
963
|
+
# 4 5 B true
|
964
|
+
# 5 6 C false
|
965
|
+
#
|
966
|
+
# puts SubFrames.new(df, [[0, 1], [2, 3, 4], [5]])
|
967
|
+
#
|
968
|
+
# # =>
|
969
|
+
# x y z
|
970
|
+
# <uint8> <string> <boolean>
|
971
|
+
# 0 1 A false
|
972
|
+
# 1 2 A true
|
973
|
+
# ---
|
974
|
+
# x y z
|
975
|
+
# <uint8> <string> <boolean>
|
976
|
+
# 0 3 B false
|
977
|
+
# 1 4 B (nil)
|
978
|
+
# 2 5 B true
|
979
|
+
# ---
|
980
|
+
# x y z
|
981
|
+
# <uint8> <string> <boolean>
|
982
|
+
# 0 6 C false
|
983
|
+
#
|
984
|
+
# @since 0.4.0
|
985
|
+
#
|
986
|
+
def to_s(limit: 16)
|
987
|
+
_to_s(limit: limit)
|
988
|
+
end
|
989
|
+
|
990
|
+
# Return summary information of self.
|
991
|
+
#
|
992
|
+
# @param limit [Integer]
|
993
|
+
# maximum number of DataFrames to show.
|
994
|
+
# @return [String]
|
995
|
+
# return class name, object id, universal DataFrame,
|
996
|
+
# size and subset sizes in a String.
|
997
|
+
# @example
|
998
|
+
# df
|
999
|
+
#
|
1000
|
+
# # =>
|
1001
|
+
# #<RedAmber::DataFrame : 6 x 3 Vectors, 0x000000000000caa8>
|
1002
|
+
# x y z
|
1003
|
+
# <uint8> <string> <boolean>
|
1004
|
+
# 0 1 A false
|
1005
|
+
# 1 2 A true
|
1006
|
+
# 2 3 B false
|
1007
|
+
# 3 4 B (nil)
|
1008
|
+
# 4 5 B true
|
1009
|
+
# 5 6 C false
|
1010
|
+
#
|
1011
|
+
# SubFrames.new(df, [[0, 1], [2, 3, 4], [5]])
|
1012
|
+
#
|
1013
|
+
# # =>
|
1014
|
+
# #<RedAmber::SubFrames : 0x000000000000c1fc>
|
1015
|
+
# @baseframe=#<RedAmber::DataFrame : 6 x 3 Vectors, 0x000000000000c170>
|
1016
|
+
# 3 SubFrames: [2, 3, 1] in sizes.
|
1017
|
+
# ---
|
1018
|
+
# #<RedAmber::DataFrame : 2 x 3 Vectors, 0x000000000002a120>
|
1019
|
+
# x y z
|
1020
|
+
# <uint8> <string> <boolean>
|
1021
|
+
# 0 1 A false
|
1022
|
+
# 1 2 A true
|
1023
|
+
# ---
|
1024
|
+
# #<RedAmber::DataFrame : 1 x 3 Vectors, 0x000000000002a134>
|
1025
|
+
# x y z
|
1026
|
+
# <uint8> <string> <boolean>
|
1027
|
+
# 0 3 B false
|
1028
|
+
# 1 4 B (nil)
|
1029
|
+
# 2 5 B true
|
1030
|
+
# ---
|
1031
|
+
# #<RedAmber::DataFrame : 1 x 3 Vectors, 0x000000000002a148>
|
1032
|
+
# x y z
|
1033
|
+
# <uint8> <string> <boolean>
|
1034
|
+
# 0 6 C false
|
1035
|
+
#
|
1036
|
+
# @since 0.4.0
|
1037
|
+
#
|
1038
|
+
def inspect(limit: 16)
|
1039
|
+
sizes_truncated = (size > limit ? sizes.take(limit) << '...' : sizes).join(', ')
|
1040
|
+
"#<#{self.class} : #{format('0x%016x', object_id)}>\n" \
|
1041
|
+
"@baseframe=#<#{baseframe.shape_str(with_id: true)}>\n" \
|
1042
|
+
"#{size} SubFrame#{pl(size)}: " \
|
1043
|
+
"[#{sizes_truncated}] in size#{pl(size)}.\n" \
|
1044
|
+
"---\n#{_to_s(limit: limit, with_id: true)}"
|
1045
|
+
end
|
1046
|
+
|
1047
|
+
private
|
1048
|
+
|
1049
|
+
def frames
|
1050
|
+
@frames ||= @enum.to_a
|
1051
|
+
end
|
1052
|
+
|
1053
|
+
def _to_s(limit: 16, with_id: false)
|
1054
|
+
a = take(limit).map do |df|
|
1055
|
+
if with_id
|
1056
|
+
"#<#{df.shape_str(with_id: with_id)}>\n" \
|
1057
|
+
"#{df.to_s(head: 2, tail: 2)}"
|
1058
|
+
else
|
1059
|
+
df.to_s(head: 2, tail: 2)
|
1060
|
+
end
|
1061
|
+
end
|
1062
|
+
a << "+ #{size - limit} more DataFrame#{pl(size - limit)}.\n" if size > limit
|
1063
|
+
a.join("---\n")
|
1064
|
+
end
|
1065
|
+
end
|
1066
|
+
end
|