isotree 0.2.2 → 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 +8 -1
- data/LICENSE.txt +2 -2
- data/README.md +32 -14
- data/ext/isotree/ext.cpp +144 -31
- data/ext/isotree/extconf.rb +7 -7
- data/lib/isotree/isolation_forest.rb +110 -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 +38 -104
- 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
@@ -0,0 +1,842 @@
|
|
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.
|
22
|
+
* "Distance approximation using Isolation Forests."
|
23
|
+
* arXiv preprint arXiv:1910.12362 (2019).
|
24
|
+
* [9] Cortes, David.
|
25
|
+
* "Imputing missing values with unsupervised random trees."
|
26
|
+
* arXiv preprint arXiv:1911.06646 (2019).
|
27
|
+
* [10] https://math.stackexchange.com/questions/3333220/expected-average-depth-in-random-binary-tree-constructed-top-to-bottom
|
28
|
+
* [11] Cortes, David.
|
29
|
+
* "Revisiting randomized choices in isolation forests."
|
30
|
+
* arXiv preprint arXiv:2110.13402 (2021).
|
31
|
+
* [12] Guha, Sudipto, et al.
|
32
|
+
* "Robust random cut forest based anomaly detection on streams."
|
33
|
+
* International conference on machine learning. PMLR, 2016.
|
34
|
+
* [13] Cortes, David.
|
35
|
+
* "Isolation forests: looking beyond tree depth."
|
36
|
+
* arXiv preprint arXiv:2111.11639 (2021).
|
37
|
+
* [14] Ting, Kai Ming, Yue Zhu, and Zhi-Hua Zhou.
|
38
|
+
* "Isolation kernel and its effect on SVM"
|
39
|
+
* Proceedings of the 24th ACM SIGKDD
|
40
|
+
* International Conference on Knowledge Discovery & Data Mining. 2018.
|
41
|
+
*
|
42
|
+
* BSD 2-Clause License
|
43
|
+
* Copyright (c) 2019-2022, David Cortes
|
44
|
+
* All rights reserved.
|
45
|
+
* Redistribution and use in source and binary forms, with or without
|
46
|
+
* modification, are permitted provided that the following conditions are met:
|
47
|
+
* * Redistributions of source code must retain the above copyright notice, this
|
48
|
+
* list of conditions and the following disclaimer.
|
49
|
+
* * Redistributions in binary form must reproduce the above copyright notice,
|
50
|
+
* this list of conditions and the following disclaimer in the documentation
|
51
|
+
* and/or other materials provided with the distribution.
|
52
|
+
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
|
53
|
+
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
54
|
+
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
|
55
|
+
* DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
|
56
|
+
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
57
|
+
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
|
58
|
+
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
|
59
|
+
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
|
60
|
+
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
61
|
+
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
62
|
+
*/
|
63
|
+
#if !defined(_FOR_R) && !defined(_FOR_PYTHON)
|
64
|
+
#include "isotree.hpp"
|
65
|
+
#include "isotree_exportable.hpp"
|
66
|
+
#include "oop_interface.hpp"
|
67
|
+
using namespace isotree;
|
68
|
+
|
69
|
+
IsolationForest::IsolationForest
|
70
|
+
(
|
71
|
+
size_t ndim, size_t ntry, CoefType coef_type, bool coef_by_prop,
|
72
|
+
bool with_replacement, bool weight_as_sample,
|
73
|
+
size_t sample_size, size_t ntrees,
|
74
|
+
size_t max_depth, size_t ncols_per_tree, bool limit_depth,
|
75
|
+
bool penalize_range, bool standardize_data,
|
76
|
+
ScoringMetric scoring_metric, bool fast_bratio, bool weigh_by_kurt,
|
77
|
+
double prob_pick_by_gain_pl, double prob_pick_by_gain_avg,
|
78
|
+
double prob_pick_by_full_gain, double prob_pick_by_dens,
|
79
|
+
double prob_pick_col_by_range, double prob_pick_col_by_var,
|
80
|
+
double prob_pick_col_by_kurt,
|
81
|
+
double min_gain, MissingAction missing_action,
|
82
|
+
CategSplit cat_split_type, NewCategAction new_cat_action,
|
83
|
+
bool all_perm, bool build_imputer, size_t min_imp_obs,
|
84
|
+
UseDepthImp depth_imp, WeighImpRows weigh_imp_rows,
|
85
|
+
uint64_t random_seed, int nthreads
|
86
|
+
)
|
87
|
+
:
|
88
|
+
ndim(ndim),
|
89
|
+
ntry(ntry),
|
90
|
+
coef_type(coef_type),
|
91
|
+
coef_by_prop(coef_by_prop),
|
92
|
+
with_replacement(with_replacement),
|
93
|
+
weight_as_sample(weight_as_sample),
|
94
|
+
sample_size(sample_size),
|
95
|
+
ntrees(ntrees),
|
96
|
+
max_depth(max_depth),
|
97
|
+
ncols_per_tree(ncols_per_tree),
|
98
|
+
limit_depth(limit_depth),
|
99
|
+
penalize_range(penalize_range),
|
100
|
+
standardize_data(standardize_data),
|
101
|
+
scoring_metric(scoring_metric),
|
102
|
+
fast_bratio(fast_bratio),
|
103
|
+
weigh_by_kurt(weigh_by_kurt),
|
104
|
+
prob_pick_by_gain_pl(prob_pick_by_gain_pl),
|
105
|
+
prob_pick_by_gain_avg(prob_pick_by_gain_avg),
|
106
|
+
prob_pick_by_full_gain(prob_pick_by_full_gain),
|
107
|
+
prob_pick_by_dens(prob_pick_by_dens),
|
108
|
+
prob_pick_col_by_range(prob_pick_col_by_range),
|
109
|
+
prob_pick_col_by_var(prob_pick_col_by_var),
|
110
|
+
prob_pick_col_by_kurt(prob_pick_col_by_kurt),
|
111
|
+
min_gain(min_gain),
|
112
|
+
missing_action(missing_action),
|
113
|
+
cat_split_type(cat_split_type),
|
114
|
+
new_cat_action(new_cat_action),
|
115
|
+
all_perm(all_perm),
|
116
|
+
build_imputer(build_imputer),
|
117
|
+
min_imp_obs(min_imp_obs),
|
118
|
+
depth_imp(depth_imp),
|
119
|
+
weigh_imp_rows(weigh_imp_rows),
|
120
|
+
random_seed(random_seed)
|
121
|
+
{}
|
122
|
+
|
123
|
+
|
124
|
+
void IsolationForest::fit(double X[], size_t nrows, size_t ncols)
|
125
|
+
{
|
126
|
+
this->check_params();
|
127
|
+
this->override_previous_fit();
|
128
|
+
|
129
|
+
auto retcode = fit_iforest(
|
130
|
+
(this->ndim == 1)? &this->model : nullptr,
|
131
|
+
(this->ndim != 1)? &this->model_ext : nullptr,
|
132
|
+
X, ncols,
|
133
|
+
(int*)nullptr, (size_t)0, (int*)nullptr,
|
134
|
+
(double*)nullptr, (int*)nullptr, (int*)nullptr,
|
135
|
+
this->ndim, this->ntry, this->coef_type, this->coef_by_prop,
|
136
|
+
(double*)nullptr, this->with_replacement, this->weight_as_sample,
|
137
|
+
nrows, this->sample_size, this->ntrees,
|
138
|
+
this->max_depth, this->ncols_per_tree,
|
139
|
+
this->limit_depth, this->penalize_range, this->standardize_data,
|
140
|
+
this->scoring_metric, this->fast_bratio,
|
141
|
+
false, (double*)nullptr,
|
142
|
+
(double*)nullptr, true,
|
143
|
+
(double*)nullptr, this->weigh_by_kurt,
|
144
|
+
this->prob_pick_by_gain_pl,
|
145
|
+
this->prob_pick_by_gain_avg,
|
146
|
+
this->prob_pick_by_full_gain,
|
147
|
+
this->prob_pick_by_dens,
|
148
|
+
this->prob_pick_col_by_range,
|
149
|
+
this->prob_pick_col_by_var,
|
150
|
+
this->prob_pick_col_by_kurt,
|
151
|
+
this->min_gain, this->missing_action,
|
152
|
+
this->cat_split_type, this->new_cat_action,
|
153
|
+
this->all_perm, &this->imputer, this->min_imp_obs,
|
154
|
+
this->depth_imp, this->weigh_imp_rows, false,
|
155
|
+
this->random_seed, false, this->nthreads
|
156
|
+
);
|
157
|
+
if (retcode != EXIT_SUCCESS) unexpected_error();
|
158
|
+
this->is_fitted = true;
|
159
|
+
}
|
160
|
+
|
161
|
+
void IsolationForest::fit(double numeric_data[], size_t ncols_numeric, size_t nrows,
|
162
|
+
int categ_data[], size_t ncols_categ, int ncat[],
|
163
|
+
double sample_weights[], double col_weights[])
|
164
|
+
{
|
165
|
+
this->check_params();
|
166
|
+
this->override_previous_fit();
|
167
|
+
|
168
|
+
auto retcode = fit_iforest(
|
169
|
+
(this->ndim == 1)? &this->model : nullptr,
|
170
|
+
(this->ndim != 1)? &this->model_ext : nullptr,
|
171
|
+
numeric_data, ncols_numeric,
|
172
|
+
categ_data, ncols_categ, ncat,
|
173
|
+
(double*)nullptr, (int*)nullptr, (int*)nullptr,
|
174
|
+
this->ndim, this->ntry, this->coef_type, this->coef_by_prop,
|
175
|
+
sample_weights, this->with_replacement, this->weight_as_sample,
|
176
|
+
nrows, this->sample_size, this->ntrees,
|
177
|
+
this->max_depth, this->ncols_per_tree,
|
178
|
+
this->limit_depth, this->penalize_range, this->standardize_data,
|
179
|
+
this->scoring_metric, this->fast_bratio,
|
180
|
+
false, (double*)nullptr,
|
181
|
+
(double*)nullptr, true,
|
182
|
+
col_weights, this->weigh_by_kurt,
|
183
|
+
this->prob_pick_by_gain_pl,
|
184
|
+
this->prob_pick_by_gain_avg,
|
185
|
+
this->prob_pick_by_full_gain,
|
186
|
+
this->prob_pick_by_dens,
|
187
|
+
this->prob_pick_col_by_range,
|
188
|
+
this->prob_pick_col_by_var,
|
189
|
+
this->prob_pick_col_by_kurt,
|
190
|
+
this->min_gain, this->missing_action,
|
191
|
+
this->cat_split_type, this->new_cat_action,
|
192
|
+
this->all_perm, &this->imputer, this->min_imp_obs,
|
193
|
+
this->depth_imp, this->weigh_imp_rows, false,
|
194
|
+
this->random_seed, false, this->nthreads
|
195
|
+
);
|
196
|
+
if (retcode != EXIT_SUCCESS) unexpected_error();
|
197
|
+
this->is_fitted = true;
|
198
|
+
}
|
199
|
+
|
200
|
+
void IsolationForest::fit(double Xc[], int Xc_ind[], int Xc_indptr[],
|
201
|
+
size_t ncols_numeric, size_t nrows,
|
202
|
+
int categ_data[], size_t ncols_categ, int ncat[],
|
203
|
+
double sample_weights[], double col_weights[])
|
204
|
+
{
|
205
|
+
this->check_params();
|
206
|
+
this->override_previous_fit();
|
207
|
+
|
208
|
+
auto retcode = fit_iforest(
|
209
|
+
(this->ndim == 1)? &this->model : nullptr,
|
210
|
+
(this->ndim != 1)? &this->model_ext : nullptr,
|
211
|
+
(double*)nullptr, ncols_numeric,
|
212
|
+
categ_data, ncols_categ, ncat,
|
213
|
+
Xc, Xc_ind, Xc_indptr,
|
214
|
+
this->ndim, this->ntry, this->coef_type, this->coef_by_prop,
|
215
|
+
sample_weights, this->with_replacement, this->weight_as_sample,
|
216
|
+
nrows, this->sample_size, this->ntrees,
|
217
|
+
this->max_depth, this->ncols_per_tree,
|
218
|
+
this->limit_depth, this->penalize_range, this->standardize_data,
|
219
|
+
this->scoring_metric, this->fast_bratio,
|
220
|
+
false, (double*)nullptr,
|
221
|
+
(double*)nullptr, true,
|
222
|
+
col_weights, this->weigh_by_kurt,
|
223
|
+
this->prob_pick_by_gain_pl,
|
224
|
+
this->prob_pick_by_gain_avg,
|
225
|
+
this->prob_pick_by_full_gain,
|
226
|
+
this->prob_pick_by_dens,
|
227
|
+
this->prob_pick_col_by_range,
|
228
|
+
this->prob_pick_col_by_var,
|
229
|
+
this->prob_pick_col_by_kurt,
|
230
|
+
this->min_gain, this->missing_action,
|
231
|
+
this->cat_split_type, this->new_cat_action,
|
232
|
+
this->all_perm, &this->imputer, this->min_imp_obs,
|
233
|
+
this->depth_imp, this->weigh_imp_rows, false,
|
234
|
+
this->random_seed, false, this->nthreads
|
235
|
+
);
|
236
|
+
if (retcode != EXIT_SUCCESS) unexpected_error();
|
237
|
+
this->is_fitted = true;
|
238
|
+
}
|
239
|
+
|
240
|
+
std::vector<double> IsolationForest::predict(double X[], size_t nrows, bool standardize)
|
241
|
+
{
|
242
|
+
this->check_is_fitted();
|
243
|
+
this->check_nthreads();
|
244
|
+
std::vector<double> out(nrows);
|
245
|
+
predict_iforest(
|
246
|
+
X, (int*)nullptr,
|
247
|
+
true, (size_t)0, (size_t)0,
|
248
|
+
(double*)nullptr, (int*)nullptr, (int*)nullptr,
|
249
|
+
(double*)nullptr, (int*)nullptr, (int*)nullptr,
|
250
|
+
nrows, this->nthreads, standardize,
|
251
|
+
(!this->model.trees.empty())? &this->model : nullptr,
|
252
|
+
(!this->model_ext.hplanes.empty())? &this->model_ext : nullptr,
|
253
|
+
out.data(), (int*)nullptr, (double*)nullptr,
|
254
|
+
(TreesIndexer*)nullptr);
|
255
|
+
return out;
|
256
|
+
}
|
257
|
+
|
258
|
+
void IsolationForest::predict(double numeric_data[], int categ_data[], bool is_col_major,
|
259
|
+
size_t nrows, size_t ld_numeric, size_t ld_categ, bool standardize,
|
260
|
+
double output_depths[], int tree_num[], double per_tree_depths[])
|
261
|
+
{
|
262
|
+
this->check_is_fitted();
|
263
|
+
this->check_nthreads();
|
264
|
+
if ((tree_num || per_tree_depths) && !this->check_can_predict_per_tree())
|
265
|
+
throw std::runtime_error("Cannot predict tree numbers/depths with this model.\n");
|
266
|
+
predict_iforest(
|
267
|
+
numeric_data, categ_data,
|
268
|
+
is_col_major, ld_numeric, ld_categ,
|
269
|
+
(double*)nullptr, (int*)nullptr, (int*)nullptr,
|
270
|
+
(double*)nullptr, (int*)nullptr, (int*)nullptr,
|
271
|
+
nrows, this->nthreads, standardize,
|
272
|
+
(!this->model.trees.empty())? &this->model : nullptr,
|
273
|
+
(!this->model_ext.hplanes.empty())? &this->model_ext : nullptr,
|
274
|
+
output_depths, tree_num, per_tree_depths,
|
275
|
+
(!this->indexer.indices.empty())? &this->indexer : nullptr);
|
276
|
+
}
|
277
|
+
|
278
|
+
void IsolationForest::predict(double X_sparse[], int X_ind[], int X_indptr[], bool is_csc,
|
279
|
+
int categ_data[], bool is_col_major, size_t ld_categ, size_t nrows, bool standardize,
|
280
|
+
double output_depths[], int tree_num[], double per_tree_depths[])
|
281
|
+
{
|
282
|
+
this->check_is_fitted();
|
283
|
+
this->check_nthreads();
|
284
|
+
if ((tree_num || per_tree_depths) && !this->check_can_predict_per_tree())
|
285
|
+
throw std::runtime_error("Cannot predict tree numbers/depths with this model.\n");
|
286
|
+
std::vector<double> out(nrows);
|
287
|
+
predict_iforest(
|
288
|
+
(double*)nullptr, categ_data,
|
289
|
+
is_col_major, (size_t)0, ld_categ,
|
290
|
+
is_csc? X_sparse : (double*)nullptr, is_csc? X_ind : (int*)nullptr, is_csc? X_indptr : (int*)nullptr,
|
291
|
+
is_csc? (double*)nullptr : X_sparse, is_csc? (int*)nullptr : X_ind, is_csc? (int*)nullptr : X_indptr,
|
292
|
+
nrows, this->nthreads, standardize,
|
293
|
+
(!this->model.trees.empty())? &this->model : nullptr,
|
294
|
+
(!this->model_ext.hplanes.empty())? &this->model_ext : nullptr,
|
295
|
+
output_depths, tree_num, per_tree_depths,
|
296
|
+
(!this->indexer.indices.empty())? &this->indexer : nullptr);
|
297
|
+
}
|
298
|
+
|
299
|
+
std::vector<double> IsolationForest::predict_distance(double X[], size_t nrows,
|
300
|
+
bool as_kernel,
|
301
|
+
bool assume_full_distr, bool standardize,
|
302
|
+
bool triangular)
|
303
|
+
{
|
304
|
+
this->check_is_fitted();
|
305
|
+
this->check_nthreads();
|
306
|
+
std::vector<double> tmat(calc_ncomb(nrows));
|
307
|
+
std::vector<double> dmat(triangular? square(nrows) : 0);
|
308
|
+
|
309
|
+
calc_similarity(X, (int*)nullptr,
|
310
|
+
(double*)nullptr, (int*)nullptr, (int*)nullptr,
|
311
|
+
nrows, false, this->nthreads, assume_full_distr, standardize, as_kernel,
|
312
|
+
(!this->model.trees.empty())? &this->model : nullptr,
|
313
|
+
(!this->model_ext.hplanes.empty())? &this->model_ext : nullptr,
|
314
|
+
tmat.data(), (double*)nullptr, (size_t)0, false,
|
315
|
+
(!this->indexer.indices.empty())? &this->indexer : nullptr,
|
316
|
+
true, (size_t)0, (size_t)0);
|
317
|
+
if (!triangular) {
|
318
|
+
double diag_filler;
|
319
|
+
if (as_kernel) {
|
320
|
+
if (standardize)
|
321
|
+
diag_filler = 1.;
|
322
|
+
else
|
323
|
+
diag_filler = std::max(this->model.trees.size(), this->model_ext.hplanes.size());
|
324
|
+
}
|
325
|
+
else {
|
326
|
+
if (standardize)
|
327
|
+
diag_filler = 0;
|
328
|
+
else
|
329
|
+
diag_filler = std::numeric_limits<double>::infinity();
|
330
|
+
}
|
331
|
+
tmat_to_dense(tmat.data(), dmat.data(), nrows, diag_filler);
|
332
|
+
}
|
333
|
+
return (triangular? tmat : dmat);
|
334
|
+
}
|
335
|
+
|
336
|
+
void IsolationForest::predict_distance(double numeric_data[], int categ_data[],
|
337
|
+
size_t nrows,
|
338
|
+
bool as_kernel,
|
339
|
+
bool assume_full_distr, bool standardize,
|
340
|
+
bool triangular,
|
341
|
+
double dist_matrix[])
|
342
|
+
{
|
343
|
+
this->check_is_fitted();
|
344
|
+
this->check_nthreads();
|
345
|
+
std::vector<double> tmat(triangular? 0 : calc_ncomb(nrows));
|
346
|
+
|
347
|
+
calc_similarity(numeric_data, categ_data,
|
348
|
+
(double*)nullptr, (int*)nullptr, (int*)nullptr,
|
349
|
+
nrows, false, this->nthreads, assume_full_distr, standardize, as_kernel,
|
350
|
+
(!this->model.trees.empty())? &this->model : nullptr,
|
351
|
+
(!this->model_ext.hplanes.empty())? &this->model_ext : nullptr,
|
352
|
+
triangular? dist_matrix : tmat.data(),
|
353
|
+
(double*)nullptr, (size_t)0, false,
|
354
|
+
(!this->indexer.indices.empty())? &this->indexer : nullptr,
|
355
|
+
true, (size_t)0, (size_t)0);
|
356
|
+
if (!triangular) {
|
357
|
+
double diag_filler;
|
358
|
+
if (as_kernel) {
|
359
|
+
if (standardize)
|
360
|
+
diag_filler = 1.;
|
361
|
+
else
|
362
|
+
diag_filler = std::max(this->model.trees.size(), this->model_ext.hplanes.size());
|
363
|
+
}
|
364
|
+
else {
|
365
|
+
if (standardize)
|
366
|
+
diag_filler = 0;
|
367
|
+
else
|
368
|
+
diag_filler = std::numeric_limits<double>::infinity();
|
369
|
+
}
|
370
|
+
tmat_to_dense(tmat.data(), dist_matrix, nrows, diag_filler);
|
371
|
+
}
|
372
|
+
}
|
373
|
+
|
374
|
+
void IsolationForest::predict_distance(double Xc[], int Xc_ind[], int Xc_indptr[], int categ_data[],
|
375
|
+
size_t nrows,
|
376
|
+
bool as_kernel,
|
377
|
+
bool assume_full_distr, bool standardize,
|
378
|
+
bool triangular,
|
379
|
+
double dist_matrix[])
|
380
|
+
{
|
381
|
+
this->check_is_fitted();
|
382
|
+
this->check_nthreads();
|
383
|
+
std::vector<double> tmat(triangular? 0 : calc_ncomb(nrows));
|
384
|
+
|
385
|
+
calc_similarity((double*)nullptr, (int*)nullptr,
|
386
|
+
Xc, Xc_ind, Xc_indptr,
|
387
|
+
nrows, false, this->nthreads, assume_full_distr, standardize, as_kernel,
|
388
|
+
(!this->model.trees.empty())? &this->model : nullptr,
|
389
|
+
(!this->model_ext.hplanes.empty())? &this->model_ext : nullptr,
|
390
|
+
triangular? dist_matrix : tmat.data(),
|
391
|
+
(double*)nullptr, (size_t)0, false,
|
392
|
+
(!this->indexer.indices.empty())? &this->indexer : nullptr,
|
393
|
+
true, (size_t)0, (size_t)0);
|
394
|
+
if (!triangular) {
|
395
|
+
double diag_filler;
|
396
|
+
if (as_kernel) {
|
397
|
+
if (standardize)
|
398
|
+
diag_filler = 1.;
|
399
|
+
else
|
400
|
+
diag_filler = std::max(this->model.trees.size(), this->model_ext.hplanes.size());
|
401
|
+
}
|
402
|
+
else {
|
403
|
+
if (standardize)
|
404
|
+
diag_filler = 0;
|
405
|
+
else
|
406
|
+
diag_filler = std::numeric_limits<double>::infinity();
|
407
|
+
}
|
408
|
+
tmat_to_dense(tmat.data(), dist_matrix, nrows, diag_filler);
|
409
|
+
}
|
410
|
+
}
|
411
|
+
|
412
|
+
void IsolationForest::impute(double X[], size_t nrows)
|
413
|
+
{
|
414
|
+
this->check_is_fitted();
|
415
|
+
this->check_nthreads();
|
416
|
+
if (this->imputer.imputer_tree.empty())
|
417
|
+
throw std::runtime_error("Model was built without imputation capabilities.\n");
|
418
|
+
impute_missing_values(X, (int*)nullptr, true,
|
419
|
+
(double*)nullptr, (int*)nullptr, (int*)nullptr,
|
420
|
+
nrows, false, this->nthreads,
|
421
|
+
(!this->model.trees.empty())? &this->model : nullptr,
|
422
|
+
(!this->model_ext.hplanes.empty())? &this->model_ext : nullptr,
|
423
|
+
this->imputer);
|
424
|
+
}
|
425
|
+
|
426
|
+
void IsolationForest::impute(double numeric_data[], int categ_data[], bool is_col_major, size_t nrows)
|
427
|
+
{
|
428
|
+
this->check_is_fitted();
|
429
|
+
if (this->imputer.imputer_tree.empty())
|
430
|
+
throw std::runtime_error("Model was built without imputation capabilities.\n");
|
431
|
+
this->check_nthreads();
|
432
|
+
impute_missing_values(numeric_data, categ_data, is_col_major,
|
433
|
+
(double*)nullptr, (int*)nullptr, (int*)nullptr,
|
434
|
+
nrows, false, this->nthreads,
|
435
|
+
(!this->model.trees.empty())? &this->model : nullptr,
|
436
|
+
(!this->model_ext.hplanes.empty())? &this->model_ext : nullptr,
|
437
|
+
this->imputer);
|
438
|
+
}
|
439
|
+
|
440
|
+
void IsolationForest::impute(double Xr[], int Xr_ind[], int Xr_indptr[],
|
441
|
+
int categ_data[], bool is_col_major, size_t nrows)
|
442
|
+
{
|
443
|
+
this->check_is_fitted();
|
444
|
+
if (this->imputer.imputer_tree.empty())
|
445
|
+
throw std::runtime_error("Model was built without imputation capabilities.\n");
|
446
|
+
this->check_nthreads();
|
447
|
+
impute_missing_values((double*)nullptr, categ_data, is_col_major,
|
448
|
+
Xr, Xr_ind, Xr_indptr,
|
449
|
+
nrows, false, this->nthreads,
|
450
|
+
(!this->model.trees.empty())? &this->model : nullptr,
|
451
|
+
(!this->model_ext.hplanes.empty())? &this->model_ext : nullptr,
|
452
|
+
this->imputer);
|
453
|
+
}
|
454
|
+
|
455
|
+
void IsolationForest::build_indexer(const bool with_distances)
|
456
|
+
{
|
457
|
+
this->check_is_fitted();
|
458
|
+
if (!this->indexer.indices.empty())
|
459
|
+
return;
|
460
|
+
if (this->missing_action == Divide)
|
461
|
+
throw std::runtime_error("Cannot build tree indexer when using 'missing_action=Divide'.\n");
|
462
|
+
if (!this->model.trees.empty() && this->new_cat_action == Weighted && this->cat_split_type == SubSet)
|
463
|
+
throw std::runtime_error("Cannot build tree indexer when using 'new_cat_action=Weighted' with single-variable model.\n");
|
464
|
+
|
465
|
+
if (!this->model.trees.empty())
|
466
|
+
build_tree_indices(this->indexer, this->model, this->nthreads, with_distances);
|
467
|
+
else if (!this->model_ext.hplanes.empty())
|
468
|
+
build_tree_indices(this->indexer, this->model_ext, this->nthreads, with_distances);
|
469
|
+
else
|
470
|
+
unexpected_error();
|
471
|
+
}
|
472
|
+
|
473
|
+
void IsolationForest::set_as_reference_points(double numeric_data[], int categ_data[], bool is_col_major,
|
474
|
+
size_t nrows, size_t ld_numeric, size_t ld_categ,
|
475
|
+
const bool with_distances)
|
476
|
+
{
|
477
|
+
this->check_is_fitted();
|
478
|
+
if (!this->model.trees.empty())
|
479
|
+
set_reference_points(&this->model, (ExtIsoForest*)NULL, &this->indexer,
|
480
|
+
with_distances,
|
481
|
+
numeric_data, categ_data,
|
482
|
+
is_col_major, ld_numeric, ld_categ,
|
483
|
+
(double*)NULL, (int*)NULL, (int*)NULL,
|
484
|
+
(double*)NULL, (int*)NULL, (int*)NULL,
|
485
|
+
nrows, this->nthreads);
|
486
|
+
else
|
487
|
+
set_reference_points((IsoForest*)NULL, &this->model_ext, &this->indexer,
|
488
|
+
with_distances,
|
489
|
+
numeric_data, categ_data,
|
490
|
+
is_col_major, ld_numeric, ld_categ,
|
491
|
+
(double*)NULL, (int*)NULL, (int*)NULL,
|
492
|
+
(double*)NULL, (int*)NULL, (int*)NULL,
|
493
|
+
nrows, this->nthreads);
|
494
|
+
}
|
495
|
+
|
496
|
+
void IsolationForest::set_as_reference_points(double Xc[], int Xc_ind[], int Xc_indptr[], int categ_data[],
|
497
|
+
size_t nrows, const bool with_distances)
|
498
|
+
{
|
499
|
+
this->check_is_fitted();
|
500
|
+
if (!this->model.trees.empty())
|
501
|
+
set_reference_points(&this->model, (ExtIsoForest*)NULL, &this->indexer,
|
502
|
+
with_distances,
|
503
|
+
(double*)NULL, (int*)NULL,
|
504
|
+
true, (size_t)0, (size_t)0,
|
505
|
+
Xc, Xc_ind, Xc_indptr,
|
506
|
+
(double*)NULL, (int*)NULL, (int*)NULL,
|
507
|
+
nrows, this->nthreads);
|
508
|
+
else
|
509
|
+
set_reference_points((IsoForest*)NULL, &this->model_ext, &this->indexer,
|
510
|
+
with_distances,
|
511
|
+
(double*)NULL, (int*)NULL,
|
512
|
+
true, (size_t)0, (size_t)0,
|
513
|
+
Xc, Xc_ind, Xc_indptr,
|
514
|
+
(double*)NULL, (int*)NULL, (int*)NULL,
|
515
|
+
nrows, this->nthreads);
|
516
|
+
}
|
517
|
+
|
518
|
+
size_t IsolationForest::get_num_reference_points() const noexcept
|
519
|
+
{
|
520
|
+
return get_number_of_reference_points(this->indexer);
|
521
|
+
}
|
522
|
+
|
523
|
+
void IsolationForest::predict_distance_to_ref_points(double numeric_data[], int categ_data[],
|
524
|
+
double Xc[], int Xc_ind[], int Xc_indptr[],
|
525
|
+
size_t nrows, bool is_col_major, size_t ld_numeric, size_t ld_categ,
|
526
|
+
bool as_kernel, bool standardize,
|
527
|
+
double dist_matrix[])
|
528
|
+
{
|
529
|
+
this->check_is_fitted();
|
530
|
+
if (this->indexer.indices.empty())
|
531
|
+
throw std::runtime_error("Model has no indexer. Cannot predict distances to indexer.\n");
|
532
|
+
if (!as_kernel && this->indexer.indices.front().node_distances.empty())
|
533
|
+
throw std::runtime_error("Model's indexer was built without distances. Cannot calculate distances to reference points.\n");
|
534
|
+
if (this->indexer.indices.front().reference_points.empty())
|
535
|
+
throw std::runtime_error("Model's indexer has no reference points. Cannot calculate distances to reference points.\n");
|
536
|
+
if (dist_matrix == NULL)
|
537
|
+
throw std::runtime_error("Passed a NULL pointer for 'dist_matrix'.\n");
|
538
|
+
|
539
|
+
calc_similarity(numeric_data, categ_data,
|
540
|
+
Xc, Xc_ind, Xc_indptr,
|
541
|
+
nrows, false, this->nthreads, true, standardize, as_kernel,
|
542
|
+
(!this->model.trees.empty())? &this->model : NULL,
|
543
|
+
(!this->model_ext.hplanes.empty())? &this->model_ext : NULL,
|
544
|
+
(double*)NULL, dist_matrix, (size_t)0, true,
|
545
|
+
&this->indexer, is_col_major, ld_numeric, ld_categ);
|
546
|
+
}
|
547
|
+
|
548
|
+
void IsolationForest::serialize(FILE *out) const
|
549
|
+
{
|
550
|
+
this->serialize_template(out);
|
551
|
+
}
|
552
|
+
|
553
|
+
void IsolationForest::serialize(std::ostream &out) const
|
554
|
+
{
|
555
|
+
this->serialize_template(out);
|
556
|
+
}
|
557
|
+
|
558
|
+
IsolationForest IsolationForest::deserialize(FILE *inp, int nthreads)
|
559
|
+
{
|
560
|
+
return deserialize_template(inp, nthreads);
|
561
|
+
}
|
562
|
+
|
563
|
+
IsolationForest IsolationForest::deserialize(std::istream &inp, int nthreads)
|
564
|
+
{
|
565
|
+
return deserialize_template(inp, nthreads);
|
566
|
+
}
|
567
|
+
|
568
|
+
std::ostream& operator<<(std::ostream &ost, const IsolationForest &model)
|
569
|
+
{
|
570
|
+
model.serialize(ost);
|
571
|
+
return ost;
|
572
|
+
}
|
573
|
+
|
574
|
+
|
575
|
+
std::ostream& isotree::operator<<(std::ostream &ost, const IsolationForest &model)
|
576
|
+
{
|
577
|
+
model.serialize(ost);
|
578
|
+
return ost;
|
579
|
+
}
|
580
|
+
|
581
|
+
std::istream& operator>>(std::istream &ist, IsolationForest &model)
|
582
|
+
{
|
583
|
+
model = IsolationForest::deserialize(ist, -1);
|
584
|
+
return ist;
|
585
|
+
}
|
586
|
+
|
587
|
+
std::istream& isotree::operator>>(std::istream &ist, IsolationForest &model)
|
588
|
+
{
|
589
|
+
model = IsolationForest::deserialize(ist, -1);
|
590
|
+
return ist;
|
591
|
+
}
|
592
|
+
|
593
|
+
IsoForest& IsolationForest::get_model()
|
594
|
+
{
|
595
|
+
if (this->ndim != 1)
|
596
|
+
throw std::runtime_error("Error: class contains an 'ExtIsoForest' model only.\n");
|
597
|
+
return this->model;
|
598
|
+
}
|
599
|
+
|
600
|
+
ExtIsoForest& IsolationForest::get_model_ext()
|
601
|
+
{
|
602
|
+
if (this->ndim == 1)
|
603
|
+
throw std::runtime_error("Error: class contains an 'IsoForest' model only.\n");
|
604
|
+
return this->model_ext;
|
605
|
+
}
|
606
|
+
|
607
|
+
Imputer& IsolationForest::get_imputer()
|
608
|
+
{
|
609
|
+
if (!this->build_imputer)
|
610
|
+
throw std::runtime_error("Error: model does not contain imputer.\n");
|
611
|
+
return this->imputer;
|
612
|
+
}
|
613
|
+
|
614
|
+
TreesIndexer& IsolationForest::get_indexer()
|
615
|
+
{
|
616
|
+
if (this->indexer.indices.empty() && (!this->model.trees.empty() || !this->model_ext.hplanes.empty()))
|
617
|
+
throw std::runtime_error("Error: model does not contain indexer.\n");
|
618
|
+
return this->indexer;
|
619
|
+
}
|
620
|
+
|
621
|
+
void IsolationForest::check_nthreads()
|
622
|
+
{
|
623
|
+
if (this->nthreads < 0) {
|
624
|
+
#ifdef _OPENMP
|
625
|
+
this->nthreads = omp_get_max_threads() + this->nthreads + 1;
|
626
|
+
#else
|
627
|
+
this->nthreads = 1;
|
628
|
+
#endif
|
629
|
+
}
|
630
|
+
if (nthreads <= 0) {
|
631
|
+
fprintf(stderr, "'isotree' got invalid 'nthreads', will set to 1.\n");
|
632
|
+
this->nthreads = 1;
|
633
|
+
}
|
634
|
+
#ifndef _OPENMP
|
635
|
+
else if (nthreads > 1) {
|
636
|
+
fprintf(stderr,
|
637
|
+
"Passed nthreads:%d to 'isotree', but library was compiled without multithreading.\n",
|
638
|
+
this->nthreads);
|
639
|
+
this->nthreads = 1;
|
640
|
+
}
|
641
|
+
#endif
|
642
|
+
}
|
643
|
+
|
644
|
+
size_t IsolationForest::get_ntrees() const
|
645
|
+
{
|
646
|
+
if (!this->model.trees.empty())
|
647
|
+
return this->model.trees.size();
|
648
|
+
else if (!this->model_ext.hplanes.empty())
|
649
|
+
return this->model_ext.hplanes.size();
|
650
|
+
else
|
651
|
+
throw std::runtime_error("Model is not fitted or is corrupted.\n");
|
652
|
+
}
|
653
|
+
|
654
|
+
bool IsolationForest::check_can_predict_per_tree() const
|
655
|
+
{
|
656
|
+
if (!this->model.trees.empty())
|
657
|
+
{
|
658
|
+
if (this->model.missing_action == Divide)
|
659
|
+
return false;
|
660
|
+
if (this->model.new_cat_action == Weighted && this->cat_split_type != SingleCateg)
|
661
|
+
{
|
662
|
+
for (const std::vector<IsoTree> &tree : this->model.trees)
|
663
|
+
for (const IsoTree &node : tree)
|
664
|
+
if (node.col_type == Categorical)
|
665
|
+
return false;
|
666
|
+
}
|
667
|
+
}
|
668
|
+
|
669
|
+
return true;
|
670
|
+
}
|
671
|
+
|
672
|
+
void IsolationForest::override_previous_fit()
|
673
|
+
{
|
674
|
+
if (this->is_fitted) {
|
675
|
+
this->model = IsoForest();
|
676
|
+
this->model_ext = ExtIsoForest();
|
677
|
+
this->imputer = Imputer();
|
678
|
+
this->indexer = TreesIndexer();
|
679
|
+
}
|
680
|
+
}
|
681
|
+
|
682
|
+
void IsolationForest::check_params()
|
683
|
+
{
|
684
|
+
this->check_nthreads();
|
685
|
+
|
686
|
+
if (this->prob_pick_by_gain_avg < 0) throw std::runtime_error("'prob_pick_by_gain_avg' must be >= 0.\n");
|
687
|
+
if (this->prob_pick_by_gain_pl < 0) throw std::runtime_error("'prob_pick_by_gain_pl' must be >= 0.\n");
|
688
|
+
if (this->prob_pick_by_full_gain < 0) throw std::runtime_error("'prob_pick_by_full_gain' must be >= 0.\n");
|
689
|
+
if (this->prob_pick_by_dens < 0) throw std::runtime_error("'prob_pick_by_dens' must be >= 0.\n");
|
690
|
+
if (this->prob_pick_col_by_range < 0) throw std::runtime_error("'prob_pick_col_by_range' must be >= 0.\n");
|
691
|
+
if (this->prob_pick_col_by_var < 0) throw std::runtime_error("'prob_pick_col_by_var' must be >= 0.\n");
|
692
|
+
if (this->prob_pick_col_by_kurt < 0) throw std::runtime_error("'prob_pick_col_by_kurt' must be >= 0.\n");
|
693
|
+
|
694
|
+
if (prob_pick_by_gain_avg + prob_pick_by_gain_pl + prob_pick_by_full_gain + prob_pick_by_dens
|
695
|
+
> 1. + 2. * std::numeric_limits<double>::epsilon())
|
696
|
+
throw std::runtime_error("Probabilities for gain-based splits sum to more than 1.\n");
|
697
|
+
|
698
|
+
if (prob_pick_col_by_var + prob_pick_col_by_var + prob_pick_col_by_kurt
|
699
|
+
> 1. + 2. * std::numeric_limits<double>::epsilon())
|
700
|
+
throw std::runtime_error("Probabilities for column choices sum to more than 1.\n");
|
701
|
+
|
702
|
+
if (min_gain < 0)
|
703
|
+
throw std::runtime_error("'min_gain' cannot be negative.\n");
|
704
|
+
|
705
|
+
if (this->ndim != 1) {
|
706
|
+
if (this->missing_action == Divide)
|
707
|
+
throw std::runtime_error("'missing_action' = 'Divide' not supported in extended model.\n");
|
708
|
+
}
|
709
|
+
|
710
|
+
if (this->coef_type != Uniform && this->coef_type != Normal)
|
711
|
+
throw std::runtime_error("Invalid 'coef_type'.\n");
|
712
|
+
if (this->missing_action != Divide && this->missing_action != Impute && this->missing_action != Fail)
|
713
|
+
throw std::runtime_error("Invalid 'missing_action'.\n");
|
714
|
+
if (this->cat_split_type != SubSet && this->cat_split_type != SingleCateg)
|
715
|
+
throw std::runtime_error("Invalid 'cat_split_type'.\n");
|
716
|
+
if (this->new_cat_action != Weighted && this->new_cat_action != Smallest && this->new_cat_action != Random)
|
717
|
+
throw std::runtime_error("Invalid 'new_cat_action'.\n");
|
718
|
+
if (this->depth_imp != Lower && this->depth_imp != Higher && this->depth_imp != Same)
|
719
|
+
throw std::runtime_error("Invalid 'depth_imp'.\n");
|
720
|
+
if (this->weigh_imp_rows != Inverse && this->weigh_imp_rows != Prop && this->weigh_imp_rows != Flat)
|
721
|
+
throw std::runtime_error("Invalid 'weigh_imp_rows'.\n");
|
722
|
+
|
723
|
+
if (this->sample_size > 0 && this->sample_size <= 2)
|
724
|
+
throw std::runtime_error("'sample_size' must be greater than 2.\n");
|
725
|
+
|
726
|
+
if (this->penalize_range && (this->scoring_metric == Density || this->scoring_metric == AdjDensity))
|
727
|
+
throw std::runtime_error("'penalize_range' is incompatible with density scoring.\n");
|
728
|
+
}
|
729
|
+
|
730
|
+
void IsolationForest::check_is_fitted() const
|
731
|
+
{
|
732
|
+
if (!this->is_fitted)
|
733
|
+
throw std::runtime_error("Model has not been fitted.\n");
|
734
|
+
}
|
735
|
+
|
736
|
+
template <class otype>
|
737
|
+
void IsolationForest::serialize_template(otype &out) const
|
738
|
+
{
|
739
|
+
this->check_is_fitted();
|
740
|
+
|
741
|
+
serialize_combined(
|
742
|
+
(!this->model.trees.empty())? &this->model : nullptr,
|
743
|
+
(!this->model_ext.hplanes.empty())? &this->model_ext : nullptr,
|
744
|
+
(!this->imputer.imputer_tree.empty())? &this->imputer : nullptr,
|
745
|
+
(!this->indexer.indices.empty())? &this->indexer : nullptr,
|
746
|
+
(char*)nullptr,
|
747
|
+
(size_t)0,
|
748
|
+
out
|
749
|
+
);
|
750
|
+
}
|
751
|
+
|
752
|
+
IsolationForest::IsolationForest(int nthreads, size_t ndim, size_t ntrees, bool build_imputer)
|
753
|
+
:
|
754
|
+
nthreads(nthreads),
|
755
|
+
ndim(ndim),
|
756
|
+
ntrees(ntrees),
|
757
|
+
build_imputer(build_imputer) {this->is_fitted = true;};
|
758
|
+
|
759
|
+
template <class itype>
|
760
|
+
IsolationForest IsolationForest::deserialize_template(itype &inp, int nthreads)
|
761
|
+
{
|
762
|
+
bool is_isotree_model = false;
|
763
|
+
bool is_compatible = false;
|
764
|
+
bool has_combined_objects = false;
|
765
|
+
bool has_IsoForest = false;
|
766
|
+
bool has_ExtIsoForest = false;
|
767
|
+
bool has_Imputer = false;
|
768
|
+
bool has_Indexer = false;
|
769
|
+
bool has_metadata = false;
|
770
|
+
size_t size_metadata = 0;
|
771
|
+
inspect_serialized_object(
|
772
|
+
inp,
|
773
|
+
is_isotree_model,
|
774
|
+
is_compatible,
|
775
|
+
has_combined_objects,
|
776
|
+
has_IsoForest,
|
777
|
+
has_ExtIsoForest,
|
778
|
+
has_Imputer,
|
779
|
+
has_Indexer,
|
780
|
+
has_metadata,
|
781
|
+
size_metadata
|
782
|
+
);
|
783
|
+
if (is_isotree_model && is_compatible && !has_combined_objects)
|
784
|
+
throw std::runtime_error("Serialized model is not compatible.\n");
|
785
|
+
|
786
|
+
IsoForest model = IsoForest();
|
787
|
+
ExtIsoForest model_ext = ExtIsoForest();
|
788
|
+
Imputer imputer = Imputer();
|
789
|
+
TreesIndexer indexer = TreesIndexer();
|
790
|
+
|
791
|
+
deserialize_combined(
|
792
|
+
inp,
|
793
|
+
&model,
|
794
|
+
&model_ext,
|
795
|
+
&imputer,
|
796
|
+
&indexer,
|
797
|
+
(char*)nullptr
|
798
|
+
);
|
799
|
+
|
800
|
+
if (model.trees.empty() && model_ext.hplanes.empty())
|
801
|
+
throw std::runtime_error("Error: model contains no trees.\n");
|
802
|
+
|
803
|
+
size_t ntrees;
|
804
|
+
size_t ndim = 3;
|
805
|
+
bool build_imputer = false;
|
806
|
+
|
807
|
+
if (!model.trees.empty()) {
|
808
|
+
ntrees = model.trees.size();
|
809
|
+
ndim = 1;
|
810
|
+
}
|
811
|
+
else {
|
812
|
+
ntrees = model_ext.hplanes.size();
|
813
|
+
}
|
814
|
+
if (!imputer.imputer_tree.empty()) {
|
815
|
+
if (imputer.imputer_tree.size() != ntrees)
|
816
|
+
throw std::runtime_error("Error: imputer has incorrect number of trees.\n");
|
817
|
+
build_imputer = true;
|
818
|
+
}
|
819
|
+
if (!indexer.indices.empty()) {
|
820
|
+
if (indexer.indices.size() != ntrees)
|
821
|
+
throw std::runtime_error("Error: indexer has incorrect number of trees.\n");
|
822
|
+
}
|
823
|
+
|
824
|
+
IsolationForest out = IsolationForest(nthreads, ndim, ntrees, build_imputer);
|
825
|
+
|
826
|
+
if (!model.trees.empty()) {
|
827
|
+
out.get_model() = std::move(model);
|
828
|
+
out.penalize_range = out.get_model().has_range_penalty;
|
829
|
+
}
|
830
|
+
else {
|
831
|
+
out.get_model_ext() = std::move(model_ext);
|
832
|
+
out.penalize_range = out.get_model_ext().has_range_penalty;
|
833
|
+
}
|
834
|
+
if (!imputer.imputer_tree.empty())
|
835
|
+
out.get_imputer() = std::move(imputer);
|
836
|
+
if (!indexer.indices.empty())
|
837
|
+
out.indexer = std::move(indexer);
|
838
|
+
|
839
|
+
return out;
|
840
|
+
}
|
841
|
+
|
842
|
+
#endif
|