isotree 0.2.0 → 0.3.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +16 -1
- data/LICENSE.txt +2 -2
- data/README.md +41 -23
- data/ext/isotree/ext.cpp +144 -31
- data/ext/isotree/extconf.rb +7 -7
- data/lib/isotree/dataset.rb +0 -1
- data/lib/isotree/isolation_forest.rb +114 -30
- data/lib/isotree/version.rb +1 -1
- data/vendor/isotree/LICENSE +1 -1
- data/vendor/isotree/README.md +165 -27
- data/vendor/isotree/include/isotree.hpp +2111 -0
- data/vendor/isotree/include/isotree_oop.hpp +394 -0
- data/vendor/isotree/inst/COPYRIGHTS +62 -0
- data/vendor/isotree/src/RcppExports.cpp +525 -52
- data/vendor/isotree/src/Rwrapper.cpp +1931 -268
- data/vendor/isotree/src/c_interface.cpp +953 -0
- data/vendor/isotree/src/crit.hpp +4232 -0
- data/vendor/isotree/src/dist.hpp +1886 -0
- data/vendor/isotree/src/exp_depth_table.hpp +134 -0
- data/vendor/isotree/src/extended.hpp +1444 -0
- data/vendor/isotree/src/external_facing_generic.hpp +399 -0
- data/vendor/isotree/src/fit_model.hpp +2401 -0
- data/vendor/isotree/src/{dealloc.cpp → headers_joined.hpp} +38 -22
- data/vendor/isotree/src/helpers_iforest.hpp +813 -0
- data/vendor/isotree/src/{impute.cpp → impute.hpp} +353 -122
- data/vendor/isotree/src/indexer.cpp +515 -0
- data/vendor/isotree/src/instantiate_template_headers.cpp +118 -0
- data/vendor/isotree/src/instantiate_template_headers.hpp +240 -0
- data/vendor/isotree/src/isoforest.hpp +1659 -0
- data/vendor/isotree/src/isotree.hpp +1804 -392
- data/vendor/isotree/src/isotree_exportable.hpp +99 -0
- data/vendor/isotree/src/merge_models.cpp +159 -16
- data/vendor/isotree/src/mult.hpp +1321 -0
- data/vendor/isotree/src/oop_interface.cpp +842 -0
- data/vendor/isotree/src/oop_interface.hpp +278 -0
- data/vendor/isotree/src/other_helpers.hpp +219 -0
- data/vendor/isotree/src/predict.hpp +1932 -0
- data/vendor/isotree/src/python_helpers.hpp +134 -0
- data/vendor/isotree/src/ref_indexer.hpp +154 -0
- data/vendor/isotree/src/robinmap/LICENSE +21 -0
- data/vendor/isotree/src/robinmap/README.md +483 -0
- data/vendor/isotree/src/robinmap/include/tsl/robin_growth_policy.h +406 -0
- data/vendor/isotree/src/robinmap/include/tsl/robin_hash.h +1620 -0
- data/vendor/isotree/src/robinmap/include/tsl/robin_map.h +807 -0
- data/vendor/isotree/src/robinmap/include/tsl/robin_set.h +660 -0
- data/vendor/isotree/src/serialize.cpp +4300 -139
- data/vendor/isotree/src/sql.cpp +141 -59
- data/vendor/isotree/src/subset_models.cpp +174 -0
- data/vendor/isotree/src/utils.hpp +3808 -0
- data/vendor/isotree/src/xoshiro.hpp +467 -0
- data/vendor/isotree/src/ziggurat.hpp +405 -0
- metadata +40 -106
- data/vendor/cereal/LICENSE +0 -24
- data/vendor/cereal/README.md +0 -85
- data/vendor/cereal/include/cereal/access.hpp +0 -351
- data/vendor/cereal/include/cereal/archives/adapters.hpp +0 -163
- data/vendor/cereal/include/cereal/archives/binary.hpp +0 -169
- data/vendor/cereal/include/cereal/archives/json.hpp +0 -1019
- data/vendor/cereal/include/cereal/archives/portable_binary.hpp +0 -334
- data/vendor/cereal/include/cereal/archives/xml.hpp +0 -956
- data/vendor/cereal/include/cereal/cereal.hpp +0 -1089
- data/vendor/cereal/include/cereal/details/helpers.hpp +0 -422
- data/vendor/cereal/include/cereal/details/polymorphic_impl.hpp +0 -796
- data/vendor/cereal/include/cereal/details/polymorphic_impl_fwd.hpp +0 -65
- data/vendor/cereal/include/cereal/details/static_object.hpp +0 -127
- data/vendor/cereal/include/cereal/details/traits.hpp +0 -1411
- data/vendor/cereal/include/cereal/details/util.hpp +0 -84
- data/vendor/cereal/include/cereal/external/base64.hpp +0 -134
- data/vendor/cereal/include/cereal/external/rapidjson/allocators.h +0 -284
- data/vendor/cereal/include/cereal/external/rapidjson/cursorstreamwrapper.h +0 -78
- data/vendor/cereal/include/cereal/external/rapidjson/document.h +0 -2652
- data/vendor/cereal/include/cereal/external/rapidjson/encodedstream.h +0 -299
- data/vendor/cereal/include/cereal/external/rapidjson/encodings.h +0 -716
- data/vendor/cereal/include/cereal/external/rapidjson/error/en.h +0 -74
- data/vendor/cereal/include/cereal/external/rapidjson/error/error.h +0 -161
- data/vendor/cereal/include/cereal/external/rapidjson/filereadstream.h +0 -99
- data/vendor/cereal/include/cereal/external/rapidjson/filewritestream.h +0 -104
- data/vendor/cereal/include/cereal/external/rapidjson/fwd.h +0 -151
- data/vendor/cereal/include/cereal/external/rapidjson/internal/biginteger.h +0 -290
- data/vendor/cereal/include/cereal/external/rapidjson/internal/diyfp.h +0 -271
- data/vendor/cereal/include/cereal/external/rapidjson/internal/dtoa.h +0 -245
- data/vendor/cereal/include/cereal/external/rapidjson/internal/ieee754.h +0 -78
- data/vendor/cereal/include/cereal/external/rapidjson/internal/itoa.h +0 -308
- data/vendor/cereal/include/cereal/external/rapidjson/internal/meta.h +0 -186
- data/vendor/cereal/include/cereal/external/rapidjson/internal/pow10.h +0 -55
- data/vendor/cereal/include/cereal/external/rapidjson/internal/regex.h +0 -740
- data/vendor/cereal/include/cereal/external/rapidjson/internal/stack.h +0 -232
- data/vendor/cereal/include/cereal/external/rapidjson/internal/strfunc.h +0 -69
- data/vendor/cereal/include/cereal/external/rapidjson/internal/strtod.h +0 -290
- data/vendor/cereal/include/cereal/external/rapidjson/internal/swap.h +0 -46
- data/vendor/cereal/include/cereal/external/rapidjson/istreamwrapper.h +0 -128
- data/vendor/cereal/include/cereal/external/rapidjson/memorybuffer.h +0 -70
- data/vendor/cereal/include/cereal/external/rapidjson/memorystream.h +0 -71
- data/vendor/cereal/include/cereal/external/rapidjson/msinttypes/inttypes.h +0 -316
- data/vendor/cereal/include/cereal/external/rapidjson/msinttypes/stdint.h +0 -300
- data/vendor/cereal/include/cereal/external/rapidjson/ostreamwrapper.h +0 -81
- data/vendor/cereal/include/cereal/external/rapidjson/pointer.h +0 -1414
- data/vendor/cereal/include/cereal/external/rapidjson/prettywriter.h +0 -277
- data/vendor/cereal/include/cereal/external/rapidjson/rapidjson.h +0 -656
- data/vendor/cereal/include/cereal/external/rapidjson/reader.h +0 -2230
- data/vendor/cereal/include/cereal/external/rapidjson/schema.h +0 -2497
- data/vendor/cereal/include/cereal/external/rapidjson/stream.h +0 -223
- data/vendor/cereal/include/cereal/external/rapidjson/stringbuffer.h +0 -121
- data/vendor/cereal/include/cereal/external/rapidjson/writer.h +0 -709
- data/vendor/cereal/include/cereal/external/rapidxml/license.txt +0 -52
- data/vendor/cereal/include/cereal/external/rapidxml/manual.html +0 -406
- data/vendor/cereal/include/cereal/external/rapidxml/rapidxml.hpp +0 -2624
- data/vendor/cereal/include/cereal/external/rapidxml/rapidxml_iterators.hpp +0 -175
- data/vendor/cereal/include/cereal/external/rapidxml/rapidxml_print.hpp +0 -428
- data/vendor/cereal/include/cereal/external/rapidxml/rapidxml_utils.hpp +0 -123
- data/vendor/cereal/include/cereal/macros.hpp +0 -154
- data/vendor/cereal/include/cereal/specialize.hpp +0 -139
- data/vendor/cereal/include/cereal/types/array.hpp +0 -79
- data/vendor/cereal/include/cereal/types/atomic.hpp +0 -55
- data/vendor/cereal/include/cereal/types/base_class.hpp +0 -203
- data/vendor/cereal/include/cereal/types/bitset.hpp +0 -176
- data/vendor/cereal/include/cereal/types/boost_variant.hpp +0 -164
- data/vendor/cereal/include/cereal/types/chrono.hpp +0 -72
- data/vendor/cereal/include/cereal/types/common.hpp +0 -129
- data/vendor/cereal/include/cereal/types/complex.hpp +0 -56
- data/vendor/cereal/include/cereal/types/concepts/pair_associative_container.hpp +0 -73
- data/vendor/cereal/include/cereal/types/deque.hpp +0 -62
- data/vendor/cereal/include/cereal/types/forward_list.hpp +0 -68
- data/vendor/cereal/include/cereal/types/functional.hpp +0 -43
- data/vendor/cereal/include/cereal/types/list.hpp +0 -62
- data/vendor/cereal/include/cereal/types/map.hpp +0 -36
- data/vendor/cereal/include/cereal/types/memory.hpp +0 -425
- data/vendor/cereal/include/cereal/types/optional.hpp +0 -66
- data/vendor/cereal/include/cereal/types/polymorphic.hpp +0 -483
- data/vendor/cereal/include/cereal/types/queue.hpp +0 -132
- data/vendor/cereal/include/cereal/types/set.hpp +0 -103
- data/vendor/cereal/include/cereal/types/stack.hpp +0 -76
- data/vendor/cereal/include/cereal/types/string.hpp +0 -61
- data/vendor/cereal/include/cereal/types/tuple.hpp +0 -123
- data/vendor/cereal/include/cereal/types/unordered_map.hpp +0 -36
- data/vendor/cereal/include/cereal/types/unordered_set.hpp +0 -99
- data/vendor/cereal/include/cereal/types/utility.hpp +0 -47
- data/vendor/cereal/include/cereal/types/valarray.hpp +0 -89
- data/vendor/cereal/include/cereal/types/variant.hpp +0 -109
- data/vendor/cereal/include/cereal/types/vector.hpp +0 -112
- data/vendor/cereal/include/cereal/version.hpp +0 -52
- data/vendor/isotree/src/Makevars +0 -4
- data/vendor/isotree/src/crit.cpp +0 -912
- data/vendor/isotree/src/dist.cpp +0 -749
- data/vendor/isotree/src/extended.cpp +0 -790
- data/vendor/isotree/src/fit_model.cpp +0 -1090
- data/vendor/isotree/src/helpers_iforest.cpp +0 -324
- data/vendor/isotree/src/isoforest.cpp +0 -771
- data/vendor/isotree/src/mult.cpp +0 -607
- data/vendor/isotree/src/predict.cpp +0 -853
- data/vendor/isotree/src/utils.cpp +0 -1566
@@ -1,771 +0,0 @@
|
|
1
|
-
/* Isolation forests and variations thereof, with adjustments for incorporation
|
2
|
-
* of categorical variables and missing values.
|
3
|
-
* Writen for C++11 standard and aimed at being used in R and Python.
|
4
|
-
*
|
5
|
-
* This library is based on the following works:
|
6
|
-
* [1] Liu, Fei Tony, Kai Ming Ting, and Zhi-Hua Zhou.
|
7
|
-
* "Isolation forest."
|
8
|
-
* 2008 Eighth IEEE International Conference on Data Mining. IEEE, 2008.
|
9
|
-
* [2] Liu, Fei Tony, Kai Ming Ting, and Zhi-Hua Zhou.
|
10
|
-
* "Isolation-based anomaly detection."
|
11
|
-
* ACM Transactions on Knowledge Discovery from Data (TKDD) 6.1 (2012): 3.
|
12
|
-
* [3] Hariri, Sahand, Matias Carrasco Kind, and Robert J. Brunner.
|
13
|
-
* "Extended Isolation Forest."
|
14
|
-
* arXiv preprint arXiv:1811.02141 (2018).
|
15
|
-
* [4] Liu, Fei Tony, Kai Ming Ting, and Zhi-Hua Zhou.
|
16
|
-
* "On detecting clustered anomalies using SCiForest."
|
17
|
-
* Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer, Berlin, Heidelberg, 2010.
|
18
|
-
* [5] https://sourceforge.net/projects/iforest/
|
19
|
-
* [6] https://math.stackexchange.com/questions/3388518/expected-number-of-paths-required-to-separate-elements-in-a-binary-tree
|
20
|
-
* [7] Quinlan, J. Ross. C4. 5: programs for machine learning. Elsevier, 2014.
|
21
|
-
* [8] Cortes, David. "Distance approximation using Isolation Forests." arXiv preprint arXiv:1910.12362 (2019).
|
22
|
-
* [9] Cortes, David. "Imputing missing values with unsupervised random trees." arXiv preprint arXiv:1911.06646 (2019).
|
23
|
-
*
|
24
|
-
* BSD 2-Clause License
|
25
|
-
* Copyright (c) 2020, David Cortes
|
26
|
-
* All rights reserved.
|
27
|
-
* Redistribution and use in source and binary forms, with or without
|
28
|
-
* modification, are permitted provided that the following conditions are met:
|
29
|
-
* * Redistributions of source code must retain the above copyright notice, this
|
30
|
-
* list of conditions and the following disclaimer.
|
31
|
-
* * Redistributions in binary form must reproduce the above copyright notice,
|
32
|
-
* this list of conditions and the following disclaimer in the documentation
|
33
|
-
* and/or other materials provided with the distribution.
|
34
|
-
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
|
35
|
-
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
36
|
-
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
|
37
|
-
* DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
|
38
|
-
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
39
|
-
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
|
40
|
-
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
|
41
|
-
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
|
42
|
-
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
43
|
-
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
44
|
-
*/
|
45
|
-
#include "isotree.hpp"
|
46
|
-
|
47
|
-
void split_itree_recursive(std::vector<IsoTree> &trees,
|
48
|
-
WorkerMemory &workspace,
|
49
|
-
InputData &input_data,
|
50
|
-
ModelParams &model_params,
|
51
|
-
std::vector<ImputeNode> *impute_nodes,
|
52
|
-
size_t curr_depth)
|
53
|
-
{
|
54
|
-
long double sum_weight = -HUGE_VAL;
|
55
|
-
|
56
|
-
/* calculate imputation statistics if desired */
|
57
|
-
if (impute_nodes != NULL)
|
58
|
-
{
|
59
|
-
if (input_data.Xc_indptr != NULL)
|
60
|
-
std::sort(workspace.ix_arr.begin() + workspace.st,
|
61
|
-
workspace.ix_arr.begin() + workspace.end + 1);
|
62
|
-
build_impute_node(impute_nodes->back(), workspace,
|
63
|
-
input_data, model_params,
|
64
|
-
*impute_nodes, curr_depth,
|
65
|
-
model_params.min_imp_obs);
|
66
|
-
}
|
67
|
-
|
68
|
-
/* check for potential isolated leafs */
|
69
|
-
if (workspace.end == workspace.st || curr_depth >= model_params.max_depth)
|
70
|
-
goto terminal_statistics;
|
71
|
-
|
72
|
-
/* with 2 observations and no weights, there's only 1 potential or assumed split */
|
73
|
-
if ((workspace.end - workspace.st) == 1 && !workspace.weights_arr.size() && !workspace.weights_map.size())
|
74
|
-
goto terminal_statistics;
|
75
|
-
|
76
|
-
/* when using weights, the split should stop when the sum of weights is <= 2 */
|
77
|
-
sum_weight = calculate_sum_weights(workspace.ix_arr, workspace.st, workspace.end, curr_depth,
|
78
|
-
workspace.weights_arr, workspace.weights_map);
|
79
|
-
|
80
|
-
if (curr_depth > 0 && (workspace.weights_arr.size() || workspace.weights_map.size()) && sum_weight < 2.5)
|
81
|
-
goto terminal_statistics;
|
82
|
-
|
83
|
-
/* for sparse matrices, need to sort the indices */
|
84
|
-
if (input_data.Xc_indptr != NULL && impute_nodes == NULL)
|
85
|
-
std::sort(workspace.ix_arr.begin() + workspace.st, workspace.ix_arr.begin() + workspace.end + 1);
|
86
|
-
|
87
|
-
/* pick column to split according to criteria */
|
88
|
-
workspace.prob_split_type = workspace.rbin(workspace.rnd_generator);
|
89
|
-
|
90
|
-
/* case1: guided, pick column and point with best gain */
|
91
|
-
if (
|
92
|
-
workspace.prob_split_type
|
93
|
-
< (
|
94
|
-
model_params.prob_pick_by_gain_avg +
|
95
|
-
model_params.prob_pick_by_gain_pl
|
96
|
-
)
|
97
|
-
)
|
98
|
-
{
|
99
|
-
workspace.determine_split = false;
|
100
|
-
|
101
|
-
/* case 1.1: column is decided by averaged gain */
|
102
|
-
if (workspace.prob_split_type < model_params.prob_pick_by_gain_avg)
|
103
|
-
workspace.criterion = Averaged;
|
104
|
-
|
105
|
-
/* case 1.2: column is decided by pooled gain */
|
106
|
-
else
|
107
|
-
workspace.criterion = Pooled;
|
108
|
-
|
109
|
-
/* evaluate gain for all columns */
|
110
|
-
trees.back().score = -HUGE_VAL; /* this is used to track the best gain */
|
111
|
-
if (input_data.Xc_indptr == NULL)
|
112
|
-
{
|
113
|
-
for (size_t col = 0; col < input_data.ncols_numeric; col++)
|
114
|
-
{
|
115
|
-
workspace.this_gain = eval_guided_crit(workspace.ix_arr.data(), workspace.st, workspace.end,
|
116
|
-
input_data.numeric_data + col * input_data.nrows,
|
117
|
-
workspace.split_ix, workspace.this_split_point,
|
118
|
-
workspace.xmin, workspace.xmax,
|
119
|
-
workspace.criterion, model_params.min_gain,
|
120
|
-
model_params.missing_action);
|
121
|
-
if (workspace.this_gain <= -HUGE_VAL)
|
122
|
-
{
|
123
|
-
workspace.cols_possible[col] = false;
|
124
|
-
}
|
125
|
-
|
126
|
-
else if (workspace.this_gain > trees.back().score)
|
127
|
-
{
|
128
|
-
trees.back().score = workspace.this_gain;
|
129
|
-
trees.back().col_num = col;
|
130
|
-
trees.back().num_split = workspace.this_split_point;
|
131
|
-
if (model_params.penalize_range)
|
132
|
-
{
|
133
|
-
trees.back().range_low = workspace.xmin - workspace.xmax + trees.back().num_split;
|
134
|
-
trees.back().range_high = workspace.xmax - workspace.xmin + trees.back().num_split;
|
135
|
-
}
|
136
|
-
}
|
137
|
-
}
|
138
|
-
|
139
|
-
}
|
140
|
-
|
141
|
-
else
|
142
|
-
{
|
143
|
-
for (size_t col = 0; col < input_data.ncols_numeric; col++)
|
144
|
-
{
|
145
|
-
workspace.this_gain = eval_guided_crit(workspace.ix_arr.data(), workspace.st, workspace.end,
|
146
|
-
col, input_data.Xc, input_data.Xc_ind, input_data.Xc_indptr,
|
147
|
-
workspace.buffer_dbl.data(), workspace.buffer_szt.data(),
|
148
|
-
workspace.this_split_point, workspace.xmin, workspace.xmax,
|
149
|
-
workspace.criterion, model_params.min_gain, model_params.missing_action);
|
150
|
-
if (workspace.this_gain <= -HUGE_VAL)
|
151
|
-
{
|
152
|
-
workspace.cols_possible[col] = false;
|
153
|
-
}
|
154
|
-
|
155
|
-
else if (workspace.this_gain > trees.back().score)
|
156
|
-
{
|
157
|
-
trees.back().score = workspace.this_gain;
|
158
|
-
trees.back().col_num = col;
|
159
|
-
trees.back().num_split = workspace.this_split_point;
|
160
|
-
if (model_params.penalize_range)
|
161
|
-
{
|
162
|
-
trees.back().range_low = workspace.xmin - workspace.xmax + trees.back().num_split;
|
163
|
-
trees.back().range_high = workspace.xmax - workspace.xmin + trees.back().num_split;
|
164
|
-
}
|
165
|
-
}
|
166
|
-
}
|
167
|
-
}
|
168
|
-
|
169
|
-
for (size_t col = 0; col < input_data.ncols_categ; col++)
|
170
|
-
{
|
171
|
-
workspace.this_gain = eval_guided_crit(workspace.ix_arr.data(), workspace.st, workspace.end,
|
172
|
-
input_data.categ_data + col * input_data.nrows, input_data.ncat[col],
|
173
|
-
workspace.buffer_szt.data(), workspace.buffer_szt.data() + input_data.max_categ,
|
174
|
-
workspace.buffer_dbl.data(), workspace.this_categ, workspace.this_split_categ.data(),
|
175
|
-
workspace.buffer_chr.data(), workspace.criterion, model_params.min_gain,
|
176
|
-
model_params.all_perm, model_params.missing_action, model_params.cat_split_type);
|
177
|
-
if (workspace.this_gain <= -HUGE_VAL)
|
178
|
-
{
|
179
|
-
workspace.cols_possible[col + input_data.ncols_numeric] = false;
|
180
|
-
}
|
181
|
-
|
182
|
-
else if (workspace.this_gain > trees.back().score)
|
183
|
-
{
|
184
|
-
trees.back().score = workspace.this_gain;
|
185
|
-
trees.back().col_num = col + input_data.ncols_numeric;
|
186
|
-
switch(model_params.cat_split_type)
|
187
|
-
{
|
188
|
-
case SingleCateg:
|
189
|
-
{
|
190
|
-
trees.back().chosen_cat = workspace.this_categ;
|
191
|
-
break;
|
192
|
-
}
|
193
|
-
|
194
|
-
case SubSet:
|
195
|
-
{
|
196
|
-
trees.back().cat_split.assign(workspace.this_split_categ.begin(),
|
197
|
-
workspace.this_split_categ.begin() + input_data.ncat[col]);
|
198
|
-
break;
|
199
|
-
}
|
200
|
-
}
|
201
|
-
}
|
202
|
-
}
|
203
|
-
|
204
|
-
|
205
|
-
if (trees.back().score <= 0.)
|
206
|
-
goto terminal_statistics;
|
207
|
-
else
|
208
|
-
trees.back().score = 0.;
|
209
|
-
|
210
|
-
if (trees.back().col_num < input_data.ncols_numeric)
|
211
|
-
{
|
212
|
-
trees.back().col_type = Numeric;
|
213
|
-
}
|
214
|
-
|
215
|
-
else
|
216
|
-
{
|
217
|
-
trees.back().col_type = Categorical;
|
218
|
-
trees.back().col_num -= input_data.ncols_numeric;
|
219
|
-
}
|
220
|
-
}
|
221
|
-
|
222
|
-
/* case2: column is chosen at random */
|
223
|
-
else
|
224
|
-
{
|
225
|
-
workspace.determine_split = true;
|
226
|
-
|
227
|
-
/* case 2.1: split point is chosen according to gain (averaged) */
|
228
|
-
if (
|
229
|
-
workspace.prob_split_type
|
230
|
-
< (
|
231
|
-
model_params.prob_pick_by_gain_avg +
|
232
|
-
model_params.prob_pick_by_gain_pl +
|
233
|
-
model_params.prob_split_by_gain_avg
|
234
|
-
)
|
235
|
-
)
|
236
|
-
workspace.criterion = Averaged;
|
237
|
-
|
238
|
-
/* case 2.2: split point is chosen according to gain (pooled) */
|
239
|
-
else if (
|
240
|
-
workspace.prob_split_type
|
241
|
-
< (
|
242
|
-
model_params.prob_pick_by_gain_avg +
|
243
|
-
model_params.prob_pick_by_gain_pl +
|
244
|
-
model_params.prob_split_by_gain_avg +
|
245
|
-
model_params.prob_split_by_gain_pl
|
246
|
-
)
|
247
|
-
)
|
248
|
-
workspace.criterion = Pooled;
|
249
|
-
|
250
|
-
/* case 2.3: split point is chosen randomly (like in the original paper) */
|
251
|
-
else
|
252
|
-
workspace.criterion = NoCrit;
|
253
|
-
|
254
|
-
|
255
|
-
/* pick column at random */
|
256
|
-
decide_column(input_data.ncols_numeric, input_data.ncols_categ,
|
257
|
-
trees.back().col_num, trees.back().col_type,
|
258
|
-
workspace.rnd_generator, workspace.runif,
|
259
|
-
workspace.col_sampler);
|
260
|
-
|
261
|
-
/* get the range of possible splits */
|
262
|
-
get_split_range(workspace, input_data, model_params, trees.back());
|
263
|
-
|
264
|
-
/* if it's not possible to split, will have to try more */
|
265
|
-
if (workspace.unsplittable)
|
266
|
-
{
|
267
|
-
/* keep track of which columns are tried */
|
268
|
-
add_unsplittable_col(workspace, trees.back(), input_data);
|
269
|
-
|
270
|
-
/* try more random columns for {(1/2) * ncols} times */
|
271
|
-
workspace.ncols_tried = 1;
|
272
|
-
do
|
273
|
-
{
|
274
|
-
decide_column(input_data.ncols_numeric, input_data.ncols_categ,
|
275
|
-
trees.back().col_num, trees.back().col_type,
|
276
|
-
workspace.rnd_generator, workspace.runif,
|
277
|
-
workspace.col_sampler);
|
278
|
-
if (!check_is_not_unsplittable_col(workspace, trees.back(), input_data))
|
279
|
-
{
|
280
|
-
get_split_range(workspace, input_data, model_params, trees.back());
|
281
|
-
if (!workspace.unsplittable)
|
282
|
-
break;
|
283
|
-
else
|
284
|
-
add_unsplittable_col(workspace, trees.back(), input_data);
|
285
|
-
}
|
286
|
-
workspace.ncols_tried++;
|
287
|
-
}
|
288
|
-
while (workspace.ncols_tried < input_data.ncols_tot / 2);
|
289
|
-
|
290
|
-
/* if that didn't work, then check all the columns that are still splittable */
|
291
|
-
if (workspace.unsplittable)
|
292
|
-
{
|
293
|
-
workspace.ncols_tried = 0; /* note: this is used here as a counter for the number of still splittable columns */
|
294
|
-
if (input_data.Xc_indptr == NULL)
|
295
|
-
{
|
296
|
-
for (size_t col = 0; col < input_data.ncols_numeric; col++)
|
297
|
-
{
|
298
|
-
if (!workspace.cols_possible[col]) continue;
|
299
|
-
get_range(workspace.ix_arr.data(), input_data.numeric_data + input_data.nrows * col,
|
300
|
-
workspace.st, workspace.end, model_params.missing_action,
|
301
|
-
workspace.xmin, workspace.xmax, workspace.unsplittable);
|
302
|
-
workspace.cols_possible[col] = !workspace.unsplittable;
|
303
|
-
workspace.ncols_tried += !workspace.unsplittable;
|
304
|
-
}
|
305
|
-
}
|
306
|
-
|
307
|
-
else
|
308
|
-
{
|
309
|
-
for (size_t col = 0; col < input_data.ncols_numeric; col++)
|
310
|
-
{
|
311
|
-
if (!workspace.cols_possible[col]) continue;
|
312
|
-
get_range(workspace.ix_arr.data(), workspace.st, workspace.end, col,
|
313
|
-
input_data.Xc, input_data.Xc_ind, input_data.Xc_indptr,
|
314
|
-
model_params.missing_action, workspace.xmin, workspace.xmax, workspace.unsplittable);
|
315
|
-
workspace.cols_possible[col] = !workspace.unsplittable;
|
316
|
-
workspace.ncols_tried += !workspace.unsplittable;
|
317
|
-
}
|
318
|
-
}
|
319
|
-
|
320
|
-
for (size_t col = 0; col < input_data.ncols_categ; col++)
|
321
|
-
{
|
322
|
-
if (!workspace.cols_possible[col + input_data.ncols_numeric]) continue;
|
323
|
-
get_categs(workspace.ix_arr.data(), input_data.categ_data + input_data.nrows * col,
|
324
|
-
workspace.st, workspace.end, input_data.ncat[col],
|
325
|
-
model_params.missing_action, workspace.categs.data(), workspace.npresent, workspace.unsplittable);
|
326
|
-
workspace.cols_possible[col + input_data.ncols_numeric] = !workspace.unsplittable;
|
327
|
-
workspace.ncols_tried += !workspace.unsplittable;
|
328
|
-
}
|
329
|
-
|
330
|
-
|
331
|
-
/* if no further splits are possible, end the procedure here */
|
332
|
-
workspace.npresent = workspace.ncols_tried;
|
333
|
-
if (!workspace.npresent) goto terminal_statistics;
|
334
|
-
|
335
|
-
/* otherwise, pick a column at random from the possible ones */
|
336
|
-
if (!workspace.col_sampler.max())
|
337
|
-
{
|
338
|
-
/* no weights by columns */
|
339
|
-
trees.back().col_num = std::uniform_int_distribution<size_t>
|
340
|
-
(0, workspace.npresent - 1)
|
341
|
-
(workspace.rnd_generator);
|
342
|
-
workspace.ncols_tried = 0;
|
343
|
-
for (size_t col = 0; col < input_data.ncols_tot; col++)
|
344
|
-
{
|
345
|
-
|
346
|
-
if (workspace.cols_possible[col])
|
347
|
-
{
|
348
|
-
if (workspace.ncols_tried == trees.back().col_num)
|
349
|
-
{
|
350
|
-
if (col < input_data.ncols_numeric)
|
351
|
-
{
|
352
|
-
trees.back().col_num = col;
|
353
|
-
trees.back().col_type = Numeric;
|
354
|
-
}
|
355
|
-
|
356
|
-
else
|
357
|
-
{
|
358
|
-
trees.back().col_num = col - input_data.ncols_numeric;
|
359
|
-
trees.back().col_type = Categorical;
|
360
|
-
}
|
361
|
-
break;
|
362
|
-
}
|
363
|
-
workspace.ncols_tried++;
|
364
|
-
}
|
365
|
-
|
366
|
-
}
|
367
|
-
}
|
368
|
-
|
369
|
-
else
|
370
|
-
{
|
371
|
-
/* weights by columns */
|
372
|
-
std::vector<double> col_weights = workspace.col_sampler.probabilities();
|
373
|
-
update_col_sampler(workspace, input_data);
|
374
|
-
|
375
|
-
decide_column(input_data.ncols_numeric, input_data.ncols_categ,
|
376
|
-
trees.back().col_num, trees.back().col_type,
|
377
|
-
workspace.rnd_generator, workspace.runif,
|
378
|
-
workspace.col_sampler);
|
379
|
-
}
|
380
|
-
|
381
|
-
}
|
382
|
-
|
383
|
-
/* finally, check the range if needed, and later decide on the split point */
|
384
|
-
if (workspace.criterion == NoCrit)
|
385
|
-
get_split_range(workspace, input_data, model_params, trees.back());
|
386
|
-
|
387
|
-
}
|
388
|
-
|
389
|
-
}
|
390
|
-
|
391
|
-
|
392
|
-
/* for numeric, choose a random point, or pick the best point as determined earlier */
|
393
|
-
if (trees.back().col_type == Numeric)
|
394
|
-
{
|
395
|
-
if (workspace.determine_split)
|
396
|
-
{
|
397
|
-
switch(workspace.criterion)
|
398
|
-
{
|
399
|
-
case NoCrit:
|
400
|
-
{
|
401
|
-
trees.back().num_split = std::uniform_real_distribution<double>
|
402
|
-
(workspace.xmin, workspace.xmax)
|
403
|
-
(workspace.rnd_generator);
|
404
|
-
break;
|
405
|
-
}
|
406
|
-
|
407
|
-
default:
|
408
|
-
{
|
409
|
-
if (input_data.Xc_indptr == NULL)
|
410
|
-
{
|
411
|
-
eval_guided_crit(workspace.ix_arr.data(), workspace.st, workspace.end,
|
412
|
-
input_data.numeric_data + trees.back().col_num * input_data.nrows,
|
413
|
-
workspace.split_ix, trees.back().num_split,
|
414
|
-
workspace.xmin, workspace.xmax,
|
415
|
-
workspace.criterion, model_params.min_gain,
|
416
|
-
model_params.missing_action);
|
417
|
-
if (model_params.missing_action == Fail) /* data is already split */
|
418
|
-
{
|
419
|
-
workspace.split_ix++;
|
420
|
-
goto follow_branches;
|
421
|
-
}
|
422
|
-
}
|
423
|
-
|
424
|
-
else
|
425
|
-
{
|
426
|
-
eval_guided_crit(workspace.ix_arr.data(), workspace.st, workspace.end,
|
427
|
-
trees.back().col_num, input_data.Xc, input_data.Xc_ind, input_data.Xc_indptr,
|
428
|
-
workspace.buffer_dbl.data(), workspace.buffer_szt.data(),
|
429
|
-
trees.back().num_split, workspace.xmin, workspace.xmax,
|
430
|
-
workspace.criterion, model_params.min_gain,
|
431
|
-
model_params.missing_action);
|
432
|
-
}
|
433
|
-
break;
|
434
|
-
}
|
435
|
-
}
|
436
|
-
|
437
|
-
if (model_params.penalize_range)
|
438
|
-
{
|
439
|
-
trees.back().range_low = workspace.xmin - workspace.xmax + trees.back().num_split;
|
440
|
-
trees.back().range_high = workspace.xmax - workspace.xmin + trees.back().num_split;
|
441
|
-
}
|
442
|
-
}
|
443
|
-
|
444
|
-
if (input_data.Xc_indptr == NULL)
|
445
|
-
divide_subset_split(workspace.ix_arr.data(), input_data.numeric_data + input_data.nrows * trees.back().col_num,
|
446
|
-
workspace.st, workspace.end, trees.back().num_split, model_params.missing_action,
|
447
|
-
workspace.st_NA, workspace.end_NA, workspace.split_ix);
|
448
|
-
else
|
449
|
-
divide_subset_split(workspace.ix_arr.data(), workspace.st, workspace.end, trees.back().col_num,
|
450
|
-
input_data.Xc, input_data.Xc_ind, input_data.Xc_indptr, trees.back().num_split,
|
451
|
-
model_params.missing_action, workspace.st_NA, workspace.end_NA, workspace.split_ix);
|
452
|
-
}
|
453
|
-
|
454
|
-
/* for categorical, there are different ways of splitting */
|
455
|
-
else
|
456
|
-
{
|
457
|
-
/* if the columns is binary, there's only one possible split */
|
458
|
-
if (input_data.ncat[trees.back().col_num] <= 2)
|
459
|
-
{
|
460
|
-
trees.back().chosen_cat = 0;
|
461
|
-
divide_subset_split(workspace.ix_arr.data(), input_data.categ_data + input_data.nrows * trees.back().col_num,
|
462
|
-
workspace.st, workspace.end, (int)0, model_params.missing_action,
|
463
|
-
workspace.st_NA, workspace.end_NA, workspace.split_ix);
|
464
|
-
trees.back().cat_split.clear();
|
465
|
-
trees.back().cat_split.shrink_to_fit();
|
466
|
-
}
|
467
|
-
|
468
|
-
/* otherwise, split according to desired type (single/subset) */
|
469
|
-
/* TODO: refactor this */
|
470
|
-
else
|
471
|
-
{
|
472
|
-
|
473
|
-
switch(model_params.cat_split_type)
|
474
|
-
{
|
475
|
-
|
476
|
-
case SingleCateg:
|
477
|
-
{
|
478
|
-
|
479
|
-
if (workspace.determine_split)
|
480
|
-
{
|
481
|
-
switch(workspace.criterion)
|
482
|
-
{
|
483
|
-
case NoCrit:
|
484
|
-
{
|
485
|
-
trees.back().chosen_cat = choose_cat_from_present(workspace, input_data, trees.back().col_num);
|
486
|
-
break;
|
487
|
-
}
|
488
|
-
|
489
|
-
default:
|
490
|
-
{
|
491
|
-
eval_guided_crit(workspace.ix_arr.data(), workspace.st, workspace.end,
|
492
|
-
input_data.categ_data + trees.back().col_num * input_data.nrows, input_data.ncat[trees.back().col_num],
|
493
|
-
workspace.buffer_szt.data(), workspace.buffer_szt.data() + input_data.max_categ,
|
494
|
-
workspace.buffer_dbl.data(), trees.back().chosen_cat, workspace.this_split_categ.data(),
|
495
|
-
workspace.buffer_chr.data(), workspace.criterion, model_params.min_gain,
|
496
|
-
model_params.all_perm, model_params.missing_action, model_params.cat_split_type);
|
497
|
-
break;
|
498
|
-
}
|
499
|
-
}
|
500
|
-
}
|
501
|
-
|
502
|
-
|
503
|
-
divide_subset_split(workspace.ix_arr.data(), input_data.categ_data + input_data.nrows * trees.back().col_num,
|
504
|
-
workspace.st, workspace.end, trees.back().chosen_cat, model_params.missing_action,
|
505
|
-
workspace.st_NA, workspace.end_NA, workspace.split_ix);
|
506
|
-
break;
|
507
|
-
}
|
508
|
-
|
509
|
-
|
510
|
-
case SubSet:
|
511
|
-
{
|
512
|
-
|
513
|
-
if (workspace.determine_split)
|
514
|
-
{
|
515
|
-
switch(workspace.criterion)
|
516
|
-
{
|
517
|
-
case NoCrit:
|
518
|
-
{
|
519
|
-
workspace.unsplittable = true;
|
520
|
-
while(workspace.unsplittable)
|
521
|
-
{
|
522
|
-
workspace.npresent = 0;
|
523
|
-
workspace.ncols_tried = 0;
|
524
|
-
for (int cat = 0; cat < input_data.ncat[trees.back().col_num]; cat++)
|
525
|
-
{
|
526
|
-
if (workspace.categs[cat] >= 0)
|
527
|
-
{
|
528
|
-
workspace.categs[cat] = workspace.rbin(workspace.rnd_generator) < 0.5;
|
529
|
-
workspace.npresent += workspace.categs[cat];
|
530
|
-
workspace.ncols_tried += !workspace.categs[cat];
|
531
|
-
}
|
532
|
-
workspace.unsplittable = !(workspace.npresent && workspace.ncols_tried);
|
533
|
-
}
|
534
|
-
}
|
535
|
-
|
536
|
-
trees.back().cat_split.assign(workspace.categs.begin(), workspace.categs.begin() + input_data.ncat[trees.back().col_num]);
|
537
|
-
break; /* NoCrit */
|
538
|
-
}
|
539
|
-
|
540
|
-
default:
|
541
|
-
{
|
542
|
-
trees.back().cat_split.resize(input_data.ncat[trees.back().col_num]);
|
543
|
-
eval_guided_crit(workspace.ix_arr.data(), workspace.st, workspace.end,
|
544
|
-
input_data.categ_data + trees.back().col_num * input_data.nrows, input_data.ncat[trees.back().col_num],
|
545
|
-
workspace.buffer_szt.data(), workspace.buffer_szt.data() + input_data.max_categ,
|
546
|
-
workspace.buffer_dbl.data(), trees.back().chosen_cat, trees.back().cat_split.data(),
|
547
|
-
workspace.buffer_chr.data(), workspace.criterion, model_params.min_gain,
|
548
|
-
model_params.all_perm, model_params.missing_action, model_params.cat_split_type);
|
549
|
-
break;
|
550
|
-
}
|
551
|
-
}
|
552
|
-
}
|
553
|
-
|
554
|
-
if (model_params.new_cat_action == Random)
|
555
|
-
for (int cat = 0; cat < input_data.ncat[trees.back().col_num]; cat++)
|
556
|
-
if (trees.back().cat_split[cat] < 0)
|
557
|
-
trees.back().cat_split[cat] = workspace.rbin(workspace.rnd_generator) < 0.5;
|
558
|
-
|
559
|
-
divide_subset_split(workspace.ix_arr.data(), input_data.categ_data + input_data.nrows * trees.back().col_num,
|
560
|
-
workspace.st, workspace.end, trees.back().cat_split.data(), model_params.missing_action,
|
561
|
-
workspace.st_NA, workspace.end_NA, workspace.split_ix);
|
562
|
-
}
|
563
|
-
|
564
|
-
}
|
565
|
-
|
566
|
-
}
|
567
|
-
|
568
|
-
}
|
569
|
-
|
570
|
-
|
571
|
-
/* if it hasn't reached the limit, continue splitting from here */
|
572
|
-
follow_branches:
|
573
|
-
{
|
574
|
-
/* add another round of separation depth for distance */
|
575
|
-
if (model_params.calc_dist && curr_depth > 0)
|
576
|
-
add_separation_step(workspace, input_data, (double)(-1));
|
577
|
-
|
578
|
-
size_t tree_from = trees.size() - 1;
|
579
|
-
size_t ix2, ix3;
|
580
|
-
std::unique_ptr<std::vector<bool>> cols_possible_ptr;
|
581
|
-
std::unique_ptr<std::discrete_distribution<size_t>> col_sampler_ptr;
|
582
|
-
trees.back().score = -1;
|
583
|
-
|
584
|
-
/* compute statistics for NAs and remember recursion indices/weights */
|
585
|
-
std::unique_ptr<RecursionState> recursion_state;
|
586
|
-
if (model_params.missing_action != Fail)
|
587
|
-
{
|
588
|
-
recursion_state = std::unique_ptr<RecursionState>(new RecursionState);
|
589
|
-
backup_recursion_state(workspace, *recursion_state);
|
590
|
-
|
591
|
-
trees.back().pct_tree_left = (long double)(workspace.st_NA - workspace.st)
|
592
|
-
/
|
593
|
-
(long double)(workspace.end - workspace.st + 1 - (workspace.end_NA - workspace.st_NA));
|
594
|
-
|
595
|
-
switch(model_params.missing_action)
|
596
|
-
{
|
597
|
-
case Impute:
|
598
|
-
{
|
599
|
-
if (trees.back().pct_tree_left >= .5)
|
600
|
-
workspace.end = workspace.end_NA - 1;
|
601
|
-
else
|
602
|
-
workspace.end = workspace.st_NA - 1;
|
603
|
-
break;
|
604
|
-
}
|
605
|
-
|
606
|
-
|
607
|
-
case Divide:
|
608
|
-
{
|
609
|
-
if (workspace.weights_map.size())
|
610
|
-
for (size_t row = workspace.st_NA; row < workspace.end_NA; row++)
|
611
|
-
workspace.weights_map[workspace.ix_arr[row]] *= trees.back().pct_tree_left;
|
612
|
-
else
|
613
|
-
for (size_t row = workspace.st_NA; row < workspace.end_NA; row++)
|
614
|
-
workspace.weights_arr[workspace.ix_arr[row]] *= trees.back().pct_tree_left;
|
615
|
-
workspace.end = workspace.end_NA - 1;
|
616
|
-
break;
|
617
|
-
}
|
618
|
-
}
|
619
|
-
}
|
620
|
-
|
621
|
-
else
|
622
|
-
{
|
623
|
-
trees.back().pct_tree_left = (long double) (workspace.split_ix - workspace.st)
|
624
|
-
/
|
625
|
-
(long double) (workspace.end - workspace.st + 1);
|
626
|
-
|
627
|
-
ix2 = workspace.split_ix;
|
628
|
-
ix3 = workspace.end;
|
629
|
-
cols_possible_ptr = std::unique_ptr<std::vector<bool>>(new std::vector<bool>);
|
630
|
-
*cols_possible_ptr = workspace.cols_possible;
|
631
|
-
if (workspace.col_sampler.max())
|
632
|
-
{
|
633
|
-
col_sampler_ptr = std::unique_ptr<std::discrete_distribution<size_t>>(new std::discrete_distribution<size_t>);
|
634
|
-
*col_sampler_ptr = workspace.col_sampler;
|
635
|
-
}
|
636
|
-
workspace.end = workspace.split_ix - 1;
|
637
|
-
}
|
638
|
-
|
639
|
-
/* Branch where to assign new categories can be pre-determined in this case */
|
640
|
-
if (
|
641
|
-
trees.back().col_type == Categorical &&
|
642
|
-
model_params.cat_split_type == SubSet &&
|
643
|
-
input_data.ncat[trees.back().col_num] > 2 &&
|
644
|
-
model_params.new_cat_action == Smallest
|
645
|
-
)
|
646
|
-
{
|
647
|
-
bool new_to_left = trees.back().pct_tree_left < 0.5;
|
648
|
-
for (int cat = 0; cat < input_data.ncat[trees.back().col_num]; cat++)
|
649
|
-
if (trees.back().cat_split[cat] < 0)
|
650
|
-
trees.back().cat_split[cat] = new_to_left;
|
651
|
-
}
|
652
|
-
|
653
|
-
/* left branch */
|
654
|
-
trees.back().tree_left = trees.size();
|
655
|
-
trees.emplace_back();
|
656
|
-
if (impute_nodes != NULL) impute_nodes->emplace_back(tree_from);
|
657
|
-
split_itree_recursive(trees,
|
658
|
-
workspace,
|
659
|
-
input_data,
|
660
|
-
model_params,
|
661
|
-
impute_nodes,
|
662
|
-
curr_depth + 1);
|
663
|
-
|
664
|
-
|
665
|
-
/* right branch */
|
666
|
-
if (model_params.missing_action != Fail)
|
667
|
-
{
|
668
|
-
restore_recursion_state(workspace, *recursion_state);
|
669
|
-
|
670
|
-
switch(model_params.missing_action)
|
671
|
-
{
|
672
|
-
case Impute:
|
673
|
-
{
|
674
|
-
if (trees[tree_from].pct_tree_left >= .5)
|
675
|
-
workspace.st = workspace.end_NA;
|
676
|
-
else
|
677
|
-
workspace.st = workspace.st_NA;
|
678
|
-
break;
|
679
|
-
}
|
680
|
-
|
681
|
-
case Divide:
|
682
|
-
{
|
683
|
-
if (workspace.weights_map.size())
|
684
|
-
for (size_t row = workspace.st_NA; row < workspace.end_NA; row++)
|
685
|
-
workspace.weights_map[workspace.ix_arr[row]] *= (1 - trees[tree_from].pct_tree_left);
|
686
|
-
else
|
687
|
-
for (size_t row = workspace.st_NA; row < workspace.end_NA; row++)
|
688
|
-
workspace.weights_arr[workspace.ix_arr[row]] *= (1 - trees[tree_from].pct_tree_left);
|
689
|
-
workspace.st = workspace.st_NA;
|
690
|
-
break;
|
691
|
-
}
|
692
|
-
}
|
693
|
-
}
|
694
|
-
|
695
|
-
else
|
696
|
-
{
|
697
|
-
workspace.st = ix2;
|
698
|
-
workspace.end = ix3;
|
699
|
-
workspace.cols_possible = std::move(*cols_possible_ptr);
|
700
|
-
if (col_sampler_ptr)
|
701
|
-
workspace.col_sampler = std::move(*col_sampler_ptr);
|
702
|
-
}
|
703
|
-
|
704
|
-
trees[tree_from].tree_right = trees.size();
|
705
|
-
trees.emplace_back();
|
706
|
-
if (impute_nodes != NULL) impute_nodes->emplace_back(tree_from);
|
707
|
-
split_itree_recursive(trees,
|
708
|
-
workspace,
|
709
|
-
input_data,
|
710
|
-
model_params,
|
711
|
-
impute_nodes,
|
712
|
-
curr_depth + 1);
|
713
|
-
}
|
714
|
-
return;
|
715
|
-
|
716
|
-
/* if it reached the limit, calculate terminal statistics */
|
717
|
-
terminal_statistics:
|
718
|
-
{
|
719
|
-
if (!workspace.weights_arr.size() && !workspace.weights_map.size())
|
720
|
-
{
|
721
|
-
trees.back().score = (double)(curr_depth + expected_avg_depth(workspace.end - workspace.st + 1));
|
722
|
-
}
|
723
|
-
|
724
|
-
else
|
725
|
-
{
|
726
|
-
if (sum_weight == -HUGE_VAL)
|
727
|
-
sum_weight = calculate_sum_weights(workspace.ix_arr, workspace.st, workspace.end, curr_depth,
|
728
|
-
workspace.weights_arr, workspace.weights_map);
|
729
|
-
trees.back().score = (double)(curr_depth + expected_avg_depth(sum_weight));
|
730
|
-
}
|
731
|
-
|
732
|
-
trees.back().cat_split.clear();
|
733
|
-
trees.back().cat_split.shrink_to_fit();
|
734
|
-
|
735
|
-
trees.back().remainder = workspace.weights_arr.size()?
|
736
|
-
sum_weight : (workspace.weights_map.size()?
|
737
|
-
sum_weight : ((double)(workspace.end - workspace.st + 1))
|
738
|
-
);
|
739
|
-
|
740
|
-
/* for distance, assume also the elements keep being split */
|
741
|
-
if (model_params.calc_dist)
|
742
|
-
add_remainder_separation_steps(workspace, input_data, sum_weight);
|
743
|
-
|
744
|
-
/* add this depth right away if requested */
|
745
|
-
if (workspace.row_depths.size())
|
746
|
-
{
|
747
|
-
if (!workspace.weights_arr.size() && !workspace.weights_map.size())
|
748
|
-
{
|
749
|
-
for (size_t row = workspace.st; row <= workspace.end; row++)
|
750
|
-
workspace.row_depths[workspace.ix_arr[row]] += trees.back().score;
|
751
|
-
}
|
752
|
-
|
753
|
-
else if (workspace.weights_arr.size())
|
754
|
-
{
|
755
|
-
for (size_t row = workspace.st; row <= workspace.end; row++)
|
756
|
-
workspace.row_depths[workspace.ix_arr[row]] += workspace.weights_arr[workspace.ix_arr[row]] * trees.back().score;
|
757
|
-
}
|
758
|
-
|
759
|
-
else
|
760
|
-
{
|
761
|
-
for (size_t row = workspace.st; row <= workspace.end; row++)
|
762
|
-
workspace.row_depths[workspace.ix_arr[row]] += workspace.weights_map[workspace.ix_arr[row]] * trees.back().score;
|
763
|
-
}
|
764
|
-
}
|
765
|
-
|
766
|
-
/* add imputations from node if requested */
|
767
|
-
if (model_params.impute_at_fit)
|
768
|
-
add_from_impute_node(impute_nodes->back(), workspace, input_data);
|
769
|
-
}
|
770
|
-
|
771
|
-
}
|