PrologMethods/Geometry/fastmks.md

+# Fast Max-Kernel Search
+
+An implementation of the single-tree and dual-tree fast max-kernel search (FastMKS) algorithm. Given a set of reference points and a set of query points, this can find the reference point with maximum kernel value for each query point
+
+# Available Predicates
+
+* [initModel/11](https://gitlab.cs.uni-duesseldorf.de/stups/abschlussarbeiten/prolog-mlpack-libary/-/wikis/PrologMethods/Geometry/fastmks#initmodel11)
+* [searchWithQuery/11](https://gitlab.cs.uni-duesseldorf.de/stups/abschlussarbeiten/prolog-mlpack-libary/-/wikis/PrologMethods/Geometry/fastmks#searchwithquery11)
+* [searchNoQuery/7](https://gitlab.cs.uni-duesseldorf.de/stups/abschlussarbeiten/prolog-mlpack-libary/-/wikis/PrologMethods/Geometry/fastmks#searchnoquery7)
+
+---
+
+[links/resources](https://gitlab.cs.uni-duesseldorf.de/stups/abschlussarbeiten/prolog-mlpack-libary/-/wikis/PrologMethods/Geometry/fastmks#connected-linksresources)
+
+## **_initModel/11_**
+
+Initializes the model on the given reference set.
+
+```prolog
+%% part of the predicate definition
+initModel(  +pointer(float_array), +integer, +integer, 
+            +string, 
+            +float32, +float32, +float32, +float32, 
+            +integer, +integer, 
+            +float32).
+```
+
+### Parameters
+| Name | Type | Description | Default |
+|------|------|-------------|---------|
+| referenceData | +matrix | The reference dataset. | - |
+| kernel | +string | Kernel type to use: "linear", "polynomial", "cosine", "gaussian", "epanechnikov", "triangular", "hyptan". | linear |
+| degree | +float | Degree of polynomial kernel. | 2.0 |
+| offset | +float | Offset of kernel (for polynomial and hyptan kernels). | 0.0 |
+| bandwidth | +float | Bandwidth (for Gaussian, Epanechnikov, and triangular kernels). | 1 |
+| scale | +float | Scale of kernel (for hyptan kernel). | 1 |
+| singleMode | +integer(bool) | If true, single-tree search is used (as opposed to dual-tree search. | (0)false |
+| naive | +integer(bool) | If true, O(n^2) naive mode is used for computation. | (0)false |
+| base | +float | Base to use during cover tree construction. | 2 |
+
+---
+
+## **_searchWithQuery/11_**
+
+Search with a different query set.
+
+```prolog
+%% part of the predicate definition
+searchWithQuery(  +pointer(float_array), +integer, +integer, 
+                  +integer, 
+                  -pointer(float_array), -integer, -integer, 
+                  -pointer(float_array), -integer, -integer, 
+                  +float32).
+```
+
+### Parameters
+| Name | Type | Description | Default |
+|------|------|-------------|---------|
+| querySet | +matrix | Set to search with. | - |
+| k | +integer | Number of maximum kernels to find. | 0 |
+| indices | -matrix | Output matrix of indices. | - |
+| kernels | -matrix | Output matrix of kernels. | - |
+| base | +float | Base to use during cover tree construction. | 2 |
+
+---
+
+## **_searchNoQuery/7_**
+
+Search with the reference set as the query set.
+
+```prolog
+%% part of the predicate definition
+searchNoQuery(  +integer, 
+                -pointer(float_array), -integer, -integer, 
+                -pointer(float_array), -integer, -integer)
+```
+
+### Parameters
+| Name | Type | Description | Default |
+|------|------|-------------|---------|
+| k | +integer | Number of maximum kernels to find. | 0 |
+| indices | -matrix | Output matrix of indices. | - |
+| kernels | -matrix | Output matrix of kernels. | - |
+
+---
+
+# Connected Links/Resources
+
+If you want a more detailed explanation, then go to the python documentation. There is most of the time a good explanation on how the methods work and what the parameters do.
+
+* [MLpack::fastmks_C++\_documentation](https://www.mlpack.org/doc/stable/doxygen/classmlpack_1_1fastmks_1_1FastMKS.html)
+* [MLpack::fastmks_Python_documentation](https://www.mlpack.org/doc/stable/python_documentation.html#fastmks)
+
+added some of the links from the python documentation
+
+* Fast max-kernel search tutorial (fastmks)
+* k-nearest-neighbor search
+* [Dual-tree Fast Exact Max-Kernel Search (pdf)](http://mlpack.org/papers/fmks.pdf)
\ No newline at end of file