1mlpack_det(1) User Commands mlpack_det(1)
2
3
4
6 mlpack_det - density estimation with density estimation trees
7
9 mlpack_det [-f int] [-m unknown] [-L int] [-l int] [-p string] [-s bool] [-T string] [-t string] [-V bool] [-M unknown] [-c string] [-g string] [-E string] [-e string] [-i string] [-h -v]
10
12 This program performs a number of functions related to Density Estima‐
13 tion Trees. The optimal Density Estimation Tree (DET) can be trained on
14 a set of data (specified by '--training_file (-t)') using cross-valida‐
15 tion (with number of folds specified with the '--folds (-f)' parame‐
16 ter). This trained density estimation tree may then be saved with the
17 '--output_model_file (-M)' output parameter.
18
19 The variable importances (that is, the feature importance values for
20 each dimension) may be saved with the '--vi_file (-i)' output parame‐
21 ter, and the density estimates for each training point may be saved
22 with the ’--training_set_estimates_file (-e)' output parameter.
23
24 Enabling path printing for each node outputs the path from the root
25 node to a leaf for each entry in the test set, or training set (if a
26 test set is not provided). Strings like 'LRLRLR' (indicating that tra‐
27 versal went to the left child, then the right child, then the left
28 child, and so forth) will be output. If 'lr-id' or 'id-lr' are given as
29 the '--path_format (-p)' parameter, then the ID (tag) of every node
30 along the path will be printed after or before the L or R character
31 indicating the direction of traversal, respectively.
32
33 This program also can provide density estimates for a set of test
34 points, specified in the '--test_file (-T)' parameter. The density
35 estimation tree used for this task will be the tree that was trained on
36 the given training points, or a tree given as the parameter
37 '--input_model_file (-m)'. The density estimates for the test points
38 may be saved using the ’--test_set_estimates_file (-E)' output parame‐
39 ter.
40
42 --folds (-f) [int]
43 The number of folds of cross-validation to perform for the esti‐
44 mation (0 is LOOCV) Default value 10.
45
46 --help (-h) [bool]
47 Default help info.
48
49 --info [string]
50 Get help on a specific module or option. Default value ''.
51
52 --input_model_file (-m) [unknown]
53 Trained density estimation tree to load. Default value ''.
54
55 --max_leaf_size (-L) [int]
56 The maximum size of a leaf in the unpruned, fully grown DET.
57 Default value 10.
58
59 --min_leaf_size (-l) [int]
60 The minimum size of a leaf in the unpruned, fully grown DET.
61 Default value 5.
62
63 --path_format (-p) [string]
64 The format of path printing: 'lr', 'id-lr', or 'lr-id'. Default
65 value 'lr'.
66
67 --skip_pruning (-s) [bool]
68 Whether to bypass the pruning process and output the unpruned
69 tree only.
70
71 --test_file (-T) [string]
72 A set of test points to estimate the density of. Default value
73 ''.
74
75 --training_file (-t) [string]
76 The data set on which to build a density estimation tree.
77 Default value ''.
78
79 --verbose (-v) [bool]
80 Display informational messages and the full list of parameters
81 and timers at the end of execution.
82
83 --version (-V) [bool]
84 Display the version of mlpack.
85
87 --output_model_file (-M) [unknown]
88 Output to save trained density estimation tree to. Default value
89 ''.
90
91 --tag_counters_file (-c) [string]
92 The file to output the number of points that went to each leaf.
93 Default value ''.
94
95 --tag_file (-g) [string]
96 The file to output the tags (and possibly paths) for each sample
97 in the test set. Default value ''.
98
99 --test_set_estimates_file (-E) [string]
100 The output estimates on the test set from the final optimally
101 pruned tree. Default value ''.
102
103 --training_set_estimates_file (-e) [string]
104 The output density estimates on the training set from the final
105 optimally pruned tree. Default value ''.
106
107 --vi_file (-i) [string]
108 The output variable importance values for each feature. Default
109 value ''.
110
112 For further information, including relevant papers, citations, and the‐
113 ory, consult the documentation found at http://www.mlpack.org or
114 included with your distribution of mlpack.
115
116
117
118mlpack-3.0.4 21 February 2019 mlpack_det(1)