1
2
3
4
5
6
7
8
9
10
11
12
13
14
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
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
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
144
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
180
181
182
183
184
185
186
187
188
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
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
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
378
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
418
419
420
|
# selected attributes of interest when generating the report for a train-/test-evaluation
VAL_ATTR_TRAIN_TEST = [ :model_uri, :training_dataset_uri, :test_dataset_uri, :prediction_feature ]
# selected attributes of interest when generating the crossvalidation report
VAL_ATTR_CV = [ :algorithm_uri, :dataset_uri, :num_folds, :crossvalidation_fold ]
# selected attributes of interest when performing classification
VAL_ATTR_CLASS = [ :percent_correct, :weighted_area_under_roc, :area_under_roc, :f_measure, :true_positive_rate, :true_negative_rate ]
VAL_ATTR_REGR = [ :root_mean_squared_error, :mean_absolute_error, :r_square ]
VAL_ATTR_BAR_PLOT_CLASS = [ :accuracy, :weighted_area_under_roc, :area_under_roc, :f_measure, :true_positive_rate, :true_negative_rate ]
VAL_ATTR_BAR_PLOT_REGR = [ :root_mean_squared_error, :mean_absolute_error, :r_square ]
# = Reports::ReportFactory
#
# creates various reports (Reports::ReportContent)
#
module Reports::ReportFactory
RT_VALIDATION = "validation"
RT_CV = "crossvalidation"
RT_ALG_COMP = "algorithm_comparison"
RT_QMRF = "qmrf"
REPORT_TYPES = [RT_VALIDATION, RT_CV, RT_ALG_COMP, RT_QMRF ]
# creates a report of a certain type according to the validation data in validation_set
#
# call-seq:
# self.create_report(type, validation_set) => Reports::ReportContent
#
def self.create_report(type, validation_set)
case type
when RT_VALIDATION
create_report_validation(validation_set)
when RT_CV
create_report_crossvalidation(validation_set)
when RT_ALG_COMP
create_report_compare_algorithms(validation_set)
else
raise "unknown report type "+type.to_s
end
end
private
def self.create_report_validation(validation_set)
raise Reports::BadRequest.new("num validations is not equal to 1") unless validation_set.size==1
val = validation_set.validations[0]
report = Reports::ReportContent.new("Validation report")
if (val.classification?)
report.add_section_result(validation_set, VAL_ATTR_TRAIN_TEST + VAL_ATTR_CLASS, "Results", "Results")
report.add_section_roc_plot(validation_set, nil, nil, "roc-plot.svg")
#val.get_prediction_feature_values.each do |class_value|
#report.add_section_roc_plot(validation_set, class_value, nil, "roc-plot-"+class_value+".svg")
#end
report.add_section_confusion_matrix(val)
else #regression
report.add_section_result(validation_set, VAL_ATTR_TRAIN_TEST + VAL_ATTR_REGR, "Results", "Results")
report.add_section_regression_plot(validation_set)
end
report.add_section_result(validation_set, Lib::ALL_PROPS, "All Results", "All Results")
report.add_section_predictions( validation_set )
return report
end
def self.create_report_crossvalidation(validation_set)
raise Reports::BadRequest.new("num validations is not >1") unless validation_set.size>1
raise Reports::BadRequest.new("crossvalidation-id not unique and != nil: "+
validation_set.get_values(:crossvalidation_id,false).inspect) if validation_set.unique_value(:crossvalidation_id)==nil
validation_set.load_cv_attributes
raise Reports::BadRequest.new("num validations ("+validation_set.size.to_s+") is not equal to num folds ("+
validation_set.unique_value(:num_folds).to_s+")") unless validation_set.unique_value(:num_folds)==validation_set.size
raise Reports::BadRequest.new("num different folds is not equal to num validations") unless validation_set.num_different_values(:crossvalidation_fold)==validation_set.size
raise Reports::BadRequest.new("validations must be either all regression, "+
+"or all classification validations") unless validation_set.all_classification? or validation_set.all_regression?
merged = validation_set.merge([:crossvalidation_id])
raise unless merged.size==1
#puts merged.get_values(:percent_correct_variance, false).inspect
report = Reports::ReportContent.new("Crossvalidation report")
if (validation_set.all_classification?)
report.add_section_result(merged, VAL_ATTR_CV+VAL_ATTR_CLASS-[:crossvalidation_fold],"Mean Results","Mean Results")
report.add_section_roc_plot(validation_set, nil, nil, "roc-plot.svg", "Roc Plot", nil, "Roc plot")
report.add_section_roc_plot(validation_set, nil, :crossvalidation_fold, "roc-plot-folds.svg", "Roc Plot", nil, "Roc plots for folds")
#validation_set.first.get_prediction_feature_values.each do |class_value|
#report.add_section_roc_plot(validation_set, class_value, nil, "roc-plot-"+class_value+".svg")
#end
report.add_section_confusion_matrix(merged.validations[0])
report.add_section_result(validation_set, VAL_ATTR_CV+VAL_ATTR_CLASS-[:num_folds], "Results","Results")
else #regression
report.add_section_result(merged, VAL_ATTR_CV+VAL_ATTR_REGR-[:crossvalidation_fold],"Mean Results","Mean Results")
report.add_section_result(validation_set, VAL_ATTR_CV+VAL_ATTR_REGR-[:num_folds], "Results","Results")
end
report.add_section_result(validation_set, Lib::ALL_PROPS, "All Results", "All Results")
report.add_section_predictions( validation_set, [:crossvalidation_fold] )
return report
end
def self.create_report_compare_algorithms(validation_set)
#validation_set.to_array([:test_dataset_uri, :model_uri, :algorithm_uri], false).each{|a| puts a.inspect}
raise Reports::BadRequest.new("num validations is not >1") unless validation_set.size>1
raise Reports::BadRequest.new("validations must be either all regression, "+
"or all classification validations") unless validation_set.all_classification? or validation_set.all_regression?
raise Reports::BadRequest.new("number of different algorithms <2: "+
validation_set.get_values(:algorithm_uri).inspect) if validation_set.num_different_values(:algorithm_uri)<2
if validation_set.has_nil_values?(:crossvalidation_id)
if validation_set.num_different_values(:test_dataset_uri)>1
# groups results into sets with equal test and training dataset
dataset_grouping = Reports::Util.group(validation_set.validations, [:test_dataset_uri, :training_dataset_uri])
# check if the same algorithms exists for each test and training dataset
Reports::Util.check_group_matching(dataset_grouping, [:algorithm_uri])
#merged = validation_set.merge([:algorithm_uri, :dataset_uri])
report = Reports::ReportContent.new("Algorithm comparison report - Many datasets")
if (validation_set.all_classification?)
report.add_section_result(validation_set,[:algorithm_uri, :test_dataset_uri]+VAL_ATTR_CLASS,"Mean Results","Mean Results")
report.add_section_ranking_plots(validation_set, :algorithm_uri, :test_dataset_uri,
[:percent_correct, :true_positive_rate, :true_negative_rate], "true")
else # regression
raise Reports::BadRequest.new("not implemented yet for regression")
end
return report
else
# this groups all validations in x different groups (arrays) according to there algorithm-uri
algorithm_grouping = Reports::Util.group(validation_set.validations, [:algorithm_uri])
# we check if there are corresponding validations in each group that have equal attributes (folds, num-folds,..)
Reports::Util.check_group_matching(algorithm_grouping, [:training_dataset_uri, :test_dataset_uri, :prediction_feature])
report = Reports::ReportContent.new("Algorithm comparison report")
if (validation_set.all_classification?)
report.add_section_bar_plot(validation_set,nil,:algorithm_uri,VAL_ATTR_BAR_PLOT_CLASS, "bar-plot.svg")
report.add_section_roc_plot(validation_set,nil, :algorithm_uri, "roc-plot.svg")
#validation_set.first.get_prediction_feature_values.each do |class_value|
#report.add_section_bar_plot(validation_set,class_value,:algorithm_uri,VAL_ATTR_CLASS, "bar-plot-"+class_value+".svg")
#report.add_section_roc_plot(validation_set, class_value, :algorithm_uri, "roc-plot-"+class_value+".svg")
#end
report.add_section_result(validation_set,[:algorithm_uri]+VAL_ATTR_CLASS,"Results","Results")
else
#regression
report.add_section_result(validation_set,[:algorithm_uri]+VAL_ATTR_REGR,"Results","Results")
report.add_section_bar_plot(validation_set,nil,:algorithm_uri,VAL_ATTR_BAR_PLOT_REGR, "bar-plot.svg")
report.add_section_regression_plot(validation_set)
#report.add_section_result(merged, VAL_ATTR_CV+VAL_ATTR_REGR-[:crossvalidation_fold],"Mean Results","Mean Results")
#report.add_section_result(validation_set, VAL_ATTR_CV+VAL_ATTR_REGR-[:num_folds], "Results","Results")
end
report.add_section_result(validation_set, Lib::ALL_PROPS, "All Results", "All Results")
return report
end
else
raise Reports::BadRequest.new("num different cross-validation-ids <2") if validation_set.num_different_values(:crossvalidation_id)<2
validation_set.load_cv_attributes
if validation_set.num_different_values(:dataset_uri)>1
# groups results into sets with equal dataset
dataset_grouping = Reports::Util.group(validation_set.validations, [:dataset_uri])
# check if equal values in each group exist
Reports::Util.check_group_matching(dataset_grouping, [:algorithm_uri, :crossvalidation_fold, :num_folds, :stratified, :random_seed])
# we only checked that equal validations exist in each dataset group, now check for each algorithm
dataset_grouping.each do |validations|
algorithm_grouping = Reports::Util.group(validations, [:algorithm_uri])
Reports::Util.check_group_matching(algorithm_grouping, [:crossvalidation_fold, :num_folds, :stratified, :random_seed])
end
merged = validation_set.merge([:algorithm_uri, :dataset_uri])
report = Reports::ReportContent.new("Algorithm comparison report - Many datasets")
if (validation_set.all_classification?)
report.add_section_result(merged,VAL_ATTR_CV+VAL_ATTR_CLASS-[:crossvalidation_fold],"Mean Results","Mean Results")
report.add_section_ranking_plots(merged, :algorithm_uri, :dataset_uri, [:acc, :auc, :sens, :spec], "true")
else # regression
report.add_section_result(merged,VAL_ATTR_CV+VAL_ATTR_REGR-[:crossvalidation_fold],"Mean Results","Mean Results")
end
return report
else
# this groups all validations in x different groups (arrays) according to there algorithm-uri
algorithm_grouping = Reports::Util.group(validation_set.validations, [:algorithm_uri])
# we check if there are corresponding validations in each group that have equal attributes (folds, num-folds,..)
Reports::Util.check_group_matching(algorithm_grouping, [:crossvalidation_fold, :num_folds, :dataset_uri, :stratified, :random_seed])
merged = validation_set.merge([:algorithm_uri])
report = Reports::ReportContent.new("Algorithm comparison report")
if (validation_set.all_classification?)
report.add_section_result(merged,VAL_ATTR_CV+VAL_ATTR_CLASS-[:crossvalidation_fold],"Mean Results","Mean Results")
true_class = validation_set.get_true_prediction_feature_value
if true_class!=nil
report.add_section_bar_plot(merged,true_class,:algorithm_uri,VAL_ATTR_BAR_PLOT_CLASS, "bar-plot.svg")
report.add_section_roc_plot(validation_set, nil, :algorithm_uri, "roc-plot.svg")
else
validation_set.get_prediction_feature_values.each do |class_value|
report.add_section_bar_plot(merged,class_value,:algorithm_uri,VAL_ATTR_BAR_PLOT_CLASS, "bar-plot-"+class_value+".svg")
report.add_section_roc_plot(validation_set, class_value, :algorithm_uri, "roc-plot-"+class_value+".svg")
end
end
report.add_section_result(validation_set,VAL_ATTR_CV+VAL_ATTR_CLASS-[:num_folds],"Results","Results")
else #regression
report.add_section_result(merged, VAL_ATTR_CV+VAL_ATTR_REGR-[:crossvalidation_fold],"Mean Results","Mean Results")
report.add_section_result(validation_set, VAL_ATTR_CV+VAL_ATTR_REGR-[:num_folds], "Results","Results")
end
return report
end
end
end
end
# = Reports::ReportContent
#
# wraps an xml-report, adds functionality for adding sections, adds a hash for tmp files
#
class Reports::ReportContent
attr_accessor :xml_report, :tmp_files
def initialize(title)
@xml_report = Reports::XMLReport.new(title, Time.now.strftime("Created at %m.%d.%Y - %H:%M"))
end
def add_section_predictions( validation_set,
validation_attributes=[],
section_title="Predictions",
section_text="This section contains predictions.",
table_title="Predictions")
section_table = @xml_report.add_section(@xml_report.get_root_element, section_title)
if validation_set.validations[0].get_predictions
@xml_report.add_paragraph(section_table, section_text) if section_text
@xml_report.add_table(section_table, table_title, Reports::PredictionUtil.predictions_to_array(validation_set, validation_attributes))
else
@xml_report.add_paragraph(section_table, "No prediction info available.")
end
end
def add_section_result( validation_set,
validation_attributes,
table_title,
section_title="Results",
section_text="This section contains results.")
section_table = @xml_report.add_section(xml_report.get_root_element, section_title)
@xml_report.add_paragraph(section_table, section_text) if section_text
vals = validation_set.to_array(validation_attributes,true,validation_set.get_true_prediction_feature_value)
vals = vals.collect{|a| a.collect{|v| v.to_s }}
#PENDING transpose values if there more than 4 columns, and there are more than columns than rows
transpose = vals[0].size>4 && vals[0].size>vals.size
@xml_report.add_table(section_table, table_title, vals, !transpose, transpose)
end
def add_section_confusion_matrix( validation,
section_title="Confusion Matrix",
section_text="This section contains the confusion matrix.",
table_title="Confusion Matrix")
section_confusion = @xml_report.add_section(xml_report.get_root_element, section_title)
@xml_report.add_paragraph(section_confusion, section_text) if section_text
@xml_report.add_table(section_confusion, table_title,
Reports::XMLReportUtil::create_confusion_matrix( validation.confusion_matrix ), false)
end
def add_section_regression_plot( validation_set,
split_set_attribute = nil,
plot_file_name="regression-plot.svg",
section_title="Regression Plot",
section_text=nil,
image_title=nil,
image_caption=nil)
section_text = "This section contains the regression plot." unless section_text
image_title = "Regression plot" unless image_title
section_regr = @xml_report.add_section(@xml_report.get_root_element, section_title)
prediction_set = validation_set.collect{ |v| v.get_predictions }
if prediction_set.size>0
section_text += "\nWARNING: regression plot information not available for all validation results" if prediction_set.size!=validation_set.size
@xml_report.add_paragraph(section_regr, section_text) if section_text
begin
plot_file_path = add_tmp_file(plot_file_name)
Reports::PlotFactory.create_regression_plot( plot_file_path, prediction_set )
@xml_report.add_imagefigure(section_regr, image_title, plot_file_name, "SVG", image_caption)
rescue RuntimeError => ex
LOGGER.error("Could not create regression plot: "+ex.message)
rm_tmp_file(plot_file_name)
@xml_report.add_paragraph(section_regr, "could not create regression plot: "+ex.message)
end
else
@xml_report.add_paragraph(section_regr, "No prediction info for regression available.")
end
end
def add_section_roc_plot( validation_set,
class_value = nil,
split_set_attribute = nil,
plot_file_name="roc-plot.svg",
section_title="Roc Plot",
section_text=nil,
image_title=nil,
image_caption=nil)
if class_value
section_text = "This section contains the roc plot for class '"+class_value+"'." unless section_text
image_title = "Roc Plot for class-value '"+class_value+"'" unless image_title
else
section_text = "This section contains the roc plot." unless section_text
image_title = "Roc Plot for all classes" unless image_title
end
section_roc = @xml_report.add_section(@xml_report.get_root_element, section_title)
prediction_set = validation_set.collect{ |v| v.get_predictions && v.get_predictions.confidence_values_available? }
if prediction_set.size>0
section_text += "\nWARNING: roc plot information not available for all validation results" if prediction_set.size!=validation_set.size
@xml_report.add_paragraph(section_roc, section_text) if section_text
begin
plot_file_path = add_tmp_file(plot_file_name)
Reports::PlotFactory.create_roc_plot( plot_file_path, prediction_set, class_value, split_set_attribute, false )#prediction_set.size>1 )
@xml_report.add_imagefigure(section_roc, image_title, plot_file_name, "SVG", image_caption)
rescue RuntimeError => ex
LOGGER.error("could not create roc plot: "+ex.message)
rm_tmp_file(plot_file_name)
@xml_report.add_paragraph(section_roc, "could not create roc plot: "+ex.message)
end
else
@xml_report.add_paragraph(section_roc, "No prediction-confidence info for roc plot available.")
end
end
def add_section_ranking_plots( validation_set,
compare_attribute,
equal_attribute,
rank_attributes,
class_value,
section_title="Ranking Plots",
section_text="This section contains the ranking plots.")
section_rank = @xml_report.add_section(@xml_report.get_root_element, section_title)
@xml_report.add_paragraph(section_rank, section_text) if section_text
rank_attributes.each{|a| add_ranking_plot(section_rank, validation_set, compare_attribute, equal_attribute, a, class_value, a.to_s+"-ranking.svg")}
end
def add_ranking_plot( report_section,
validation_set,
compare_attribute,
equal_attribute,
rank_attribute,
class_value=nil,
plot_file_name="ranking.svg",
image_title=nil,
image_caption=nil)
image_title = "Ranking Plot for class value: '"+class_value.to_s+"'" if image_title==nil
plot_file_path = add_tmp_file(plot_file_name)
Reports::PlotFactory::create_ranking_plot(plot_file_path, validation_set, compare_attribute, equal_attribute, rank_attribute, class_value)
@xml_report.add_imagefigure(report_section, image_title, plot_file_name, "SVG", image_caption)
end
def add_section_bar_plot(validation_set,
class_value,
title_attribute,
value_attributes,
plot_file_name="bar-plot.svg",
section_title="Bar Plot",
section_text=nil,
image_title=nil,
image_caption=nil)
if class_value
section_text = "This section contains the bar plot for class '"+class_value+"'." unless section_text
image_title = "Bar Plot for class-value '"+class_value+"'" unless image_title
else
section_text = "This section contains the bar plot." unless section_text
image_title = "Bar Plot for all classes" unless image_title
end
section_bar = @xml_report.add_section(@xml_report.get_root_element, section_title)
@xml_report.add_paragraph(section_bar, section_text) if section_text
plot_file_path = add_tmp_file(plot_file_name)
Reports::PlotFactory.create_bar_plot(plot_file_path, validation_set, class_value, title_attribute, value_attributes )
@xml_report.add_imagefigure(section_bar, image_title, plot_file_name, "SVG", image_caption)
end
private
def add_tmp_file(tmp_file_name)
@tmp_files = {} unless @tmp_files
raise "file name already exits" if @tmp_files[tmp_file_name] || (@text_files && @text_files[tmp_file_name])
tmp_file_path = Reports::Util.create_tmp_file(tmp_file_name)
@tmp_files[tmp_file_name] = tmp_file_path
return tmp_file_path
end
def rm_tmp_file(tmp_file_name)
@tmp_files.delete(tmp_file_name) if @tmp_files.has_key?(tmp_file_name)
end
end
|