MKLpy-master
所属分类:Windows编程
开发工具:Python
文件大小:44KB
下载次数:11
上传日期:2019-05-20 21:29:55
上 传 者:
the混
说明: 一个多核学习python版本的代码,里面有平均核方法
(A python library of multiple kernel learning)
文件列表:
LICENSE (35149, 2019-01-19)
MKLpy (0, 2019-01-19)
MKLpy\__init__.py (2083, 2019-01-19)
MKLpy\algorithms (0, 2019-01-19)
MKLpy\algorithms\EasyMKL.py (3094, 2019-01-19)
MKLpy\algorithms\__init__.py (242, 2019-01-19)
MKLpy\algorithms\base.py (4525, 2019-01-19)
MKLpy\algorithms\komd.py (10664, 2019-01-19)
MKLpy\arrange.py (2982, 2019-01-19)
MKLpy\lists (0, 2019-01-19)
MKLpy\lists\__init__.py (312, 2019-01-19)
MKLpy\lists\generator.py (818, 2019-01-19)
MKLpy\lists\kernel_list.py (1608, 2019-01-19)
MKLpy\metrics (0, 2019-01-19)
MKLpy\metrics\__init__.py (1029, 2019-01-19)
MKLpy\metrics\alignment.py (2184, 2019-01-19)
MKLpy\metrics\evaluate.py (3801, 2019-01-19)
MKLpy\metrics\pairwise.py (2642, 2019-01-19)
MKLpy\model_selection (0, 2019-01-19)
MKLpy\model_selection\__init__.py (202, 2019-01-19)
MKLpy\model_selection\splits.py (476, 2019-01-19)
MKLpy\model_selection\validation.py (1273, 2019-01-19)
MKLpy\multiclass.py (7572, 2019-01-19)
MKLpy\preprocessing (0, 2019-01-19)
MKLpy\preprocessing\__init__.py (299, 2019-01-19)
MKLpy\preprocessing\binarization.py (6033, 2019-01-19)
MKLpy\preprocessing\data_preprocessing.py (2222, 2019-01-19)
MKLpy\preprocessing\kernel_preprocessing.py (1788, 2019-01-19)
MKLpy\utils (0, 2019-01-19)
MKLpy\utils\__init__.py (161, 2019-01-19)
MKLpy\utils\exceptions.py (626, 2019-01-19)
MKLpy\utils\matrices.py (748, 2019-01-19)
MKLpy\utils\mkl_checks.py (1721, 2019-01-19)
MKLpy\utils\validation.py (2540, 2019-01-19)
MKLpy\weights.py (1795, 2019-01-19)
examples (0, 2019-01-19)
examples\test_modelselection.py (2786, 2019-01-19)
... ...
MKLpy
=====
MKLpy is a framework for Multiple Kernel Learning and kernel machines scikit-compliant.
This package contains:
* MKL algorithms
* EasyMKL
* Average of kernels
* Soon available: GRAM, MEMO, SimpleMKL
* tools to operate on kernels, such as normalization, centering, summation, mean...;
* metrics, such as kernel_alignment, radius, margin, spectral ratio...;
* kernel functions, such as homogeneous polynomial and boolean kernels (disjunctive, conjunctive, DNF, CNF).
The 'examples' folder contains useful snippets of code.
For more informations about classification, kernels and predictors visit `Link scikit-learn `_
requirements
------------
To work properly, MKLpy requires:
* numpy
* scikit-learn (v. 0.20.0)
* cvxopt
近期下载者:
相关文件:
收藏者: