1mlpack_fastmks(1) General Commands Manual mlpack_fastmks(1)
2
3
4
6 mlpack_fastmks - fastmks (fast max-kernel search)
7
9 mlpack_fastmks [-h] [-v]
10
12 This program will find the k maximum kernel of a set of points, using a
13 query set and a reference set (which can optionally be the same set).
14 More specifically, for each point in the query set, the k points in the
15 reference set with maximum kernel evaluations are found. The kernel
16 function used is specified by --kernel.
17
18 For example, the following command will calculate, for each point in
19 ’query.csv', the five points in 'reference.csv' with maximum kernel
20 evaluation using the linear kernel. The kernel evaluations are stored
21 in 'kernels.csv' and the indices are stored in 'indices.csv'.
22
23 $ fastmks --k 5 --reference_file reference.csv --query_file query.csv
24 --indices_file indices.csv --kernels_file kernels.csv --kernel linear
25
26 The output files are organized such that row i and column j in the
27 indices output file corresponds to the index of the point in the refer‐
28 ence set that has i'th largest kernel evaluation with the point in the
29 query set with index j. Row i and column j in the kernels output file
30 corresponds to the kernel evaluation between those two points.
31
32 This executable performs FastMKS using a cover tree. The base used to
33 build the cover tree can be specified with the --base option.
34
36 --bandwidth (-w) [double]
37 Bandwidth (for Gaussian, Epanechnikov, and triangular kernels).
38 Default value 1.
39
40 --base (-b) [double]
41 Base to use during cover tree construction. Default value 2.
42
43 --degree (-d) [double]
44 Degree of polynomial kernel. Default value 2.
45
46 --help (-h)
47 Default help info.
48
49 --info [string]
50 Get help on a specific module or option. Default value ''.
51 --input_model_file (-m) [string] File containing FastMKS model.
52 Default value ’'.
53
54 --k (-k) [int]
55 Number of maximum kernels to find. Default value 0.
56
57 --kernel (-K) [string]
58 Kernel type to use: 'linear', 'polynomial', ’cosine', 'gauss‐
59 ian', 'epanechnikov', ’triangular', 'hyptan'. Default value
60 ’linear'.
61
62 --naive (-N)
63 If true, O(n^2) naive mode is used for computation.
64
65 --offset (-o) [double]
66 Offset of kernel (for polynomial and hyptan kernels). Default
67 value 0.
68
69 --query_file (-q) [string]
70 File containing the query dataset. Default value ''. --refer‐
71 ence_file (-r) [string] File containing the reference dataset.
72 Default value ''.
73
74 --scale (-s) [double]
75 Scale of kernel (for hyptan kernel). Default value 1.
76
77 --single (-S)
78 If true, single-tree search is used (as opposed to dual-tree
79 search.
80
81 --verbose (-v)
82 Display informational messages and the full list of parameters
83 and timers at the end of execution.
84
85 --version (-V)
86 Display the version of mlpack.
87
89 --indices_file (-i) [string] File to save indices of kernels into.
90 Default value ''. --kernels_file (-p) [string] File to save kernels
91 into. Default value ''. --output_model_file (-M) [string] File to save
92 FastMKS model to. Default value ’'.
93
96 For further information, including relevant papers, citations, and the‐
97 ory, For further information, including relevant papers, citations, and
98 theory, consult the documentation found at http://www.mlpack.org or
99 included with your consult the documentation found at
100 http://www.mlpack.org or included with your DISTRIBUTION OF MLPACK.
101 DISTRIBUTION OF MLPACK.
102
103
104
105 mlpack_fastmks(1)