Submit
Path:
~
/
/
proc
/
thread-self
/
root
/
opt
/
alt
/
python35
/
lib64
/
python3.5
/
site-packages
/
sklearn
/
utils
/
__pycache__
/
File Content:
linear_assignment_.cpython-35.pyc
��(X4% � @ s� d Z d d l Z d d l m Z d d � Z Gd d � d e � Z d d � Z d d � Z d d � Z d d � Z d d � Z d d � Z d S)zs Solve the unique lowest-cost assignment problem using the Hungarian algorithm (also known as Munkres algorithm). � N� )�astypec C s> t | � j � } | j � t j | d t �} d | _ | S)a5 Solve the linear assignment problem using the Hungarian algorithm. The problem is also known as maximum weight matching in bipartite graphs. The method is also known as the Munkres or Kuhn-Munkres algorithm. Parameters ---------- X : array The cost matrix of the bipartite graph Returns ------- indices : array, The pairs of (row, col) indices in the original array giving the original ordering. References ---------- 1. http://www.public.iastate.edu/~ddoty/HungarianAlgorithm.html 2. Harold W. Kuhn. The Hungarian Method for the assignment problem. *Naval Research Logistics Quarterly*, 2:83-97, 1955. 3. Harold W. Kuhn. Variants of the Hungarian method for assignment problems. *Naval Research Logistics Quarterly*, 3: 253-258, 1956. 4. Munkres, J. Algorithms for the Assignment and Transportation Problems. *Journal of the Society of Industrial and Applied Mathematics*, 5(1):32-38, March, 1957. 5. https://en.wikipedia.org/wiki/Hungarian_algorithm �dtyper � ���)r r )� _hungarian�tolist�sort�np�array�int�shape)�X�indices� r �/linear_assignment_.py�linear_assignment s " r c @ s: e Z d Z d Z d d � Z d d � Z d d � Z d S) �_HungarianStatez�State of one execution of the Hungarian algorithm. Parameters ---------- cost_matrix : 2D matrix The cost matrix. Does not have to be square. c C s� t j | � } | j d | j d k } | rD | j j � | _ n | j � | _ | | _ | j j \ } } t j | d t j �| _ t j | d t j �| _ d | _ d | _ t j | | d f d t �| _ t j | | f d t �| _ d S)Nr r r r )r Z atleast_2dr �T�copy�C� transposedZones�bool� row_uncovered� col_uncovered�Z0_r�Z0_cZzerosr �path�marked)�self� cost_matrixr �n�mr r r �__init__J s "z_HungarianState.__init__c C s? t j | j | d k � } | j | | f d k r; d } | S)z� Find the first prime element in the specified row. Returns the column index, or -1 if no starred element was found. r r r )r �argmaxr )r �row�colr r r �_find_prime_in_rowa s z"_HungarianState._find_prime_in_rowc C s* d | j d d � <d | j d d � <d S)z Clear all covered matrix cellsTN)r r )r r r r � _clear_coversk s z _HungarianState._clear_coversN)�__name__� __module__� __qualname__�__doc__r# r' r( r r r r r A s r c C s� t | � } d | j k r! d n t } x | d k rE | | � } q* Wt j t j | j d k � � j } | j r� | d d � d d d � f } | S)a� The Hungarian algorithm. Calculate the Munkres solution to the classical assignment problem and return the indices for the lowest-cost pairings. Parameters ---------- cost_matrix : 2D matrix The cost matrix. Does not have to be square. Returns ------- indices : 2D array of indices The pairs of (row, col) indices in the original array giving the original ordering. r Nr r ) r r �_step1r r �wherer r r )r �state�step�resultsr r r r q s $ r c C s� | j | j j d d � d d � t j f 8_ xs t t j | j d k � � D]S \ } } | j | rP | j | rP d | j | | f <d | j | <d | j | <qP W| j � t S)z$Steps 1 and 2 in the Wikipedia page.�axisr Nr F) r �minr �newaxis�zipr. r r r r( �_step3)r/ �i�jr r r r- � s 1+ r- c C sO | j d k } d | j t j | d d �<| j � | j j d k rK t Sd S)z� Cover each column containing a starred zero. If n columns are covered, the starred zeros describe a complete set of unique assignments. In this case, Go to DONE, otherwise, Go to Step 4. r Fr2 r N)r r r �any�sumr r �_step4)r/ r r r r r6 � s r6 c C s� | j d k j t j � } | | j d d � t j f } | t | j d t j d d �9} | j j d } | j j d } xt j t j | � | | f � \ } } | | | f d k r� t Sd | j | | f <t j | j | d k � } | j | | f d k s | | _ | | _ t S| } d | j | <d | j | <| d d � | f t | j d t j d d �| d d � | f <d | | <q� Wd S) aS Find a noncovered zero and prime it. If there is no starred zero in the row containing this primed zero, Go to Step 5. Otherwise, cover this row and uncover the column containing the starred zero. Continue in this manner until there are no uncovered zeros left. Save the smallest uncovered value and Go to Step 6. r Nr r Fr r T)r r r r r r4 r r Z unravel_indexr$ �_step6r r r �_step5)r/ r Z covered_Cr! r" r% r& Zstar_colr r r r; � s* "' /r; c C s d } | j } | j | | d f <| j | | d f <xt j | j d d � | | d f f d k � } | j | | | d f f d k s� Pn8 | d 7} | | | d f <| | d d f | | d f <t j | j | | d f d k � } | j | | f d k rd } | d 7} | | d d f | | d f <| | | d f <q8 Wx� t | d � D]� } | j | | d f | | d f f d k r�d | j | | d f | | d f f <q[d | j | | d f | | d f f <q[W| j � d | j | j d k <t S)a# Construct a series of alternating primed and starred zeros as follows. Let Z0 represent the uncovered primed zero found in Step 4. Let Z1 denote the starred zero in the column of Z0 (if any). Let Z2 denote the primed zero in the row of Z1 (there will always be one). Continue until the series terminates at a primed zero that has no starred zero in its column. Unstar each starred zero of the series, star each primed zero of the series, erase all primes and uncover every line in the matrix. Return to Step 3 r r Nr r ) r r r r r$ r �ranger( r6 )r/ �countr r% r& r7 r r r r= � s0 2# & -*+ r= c C s� t j | j � r� t j | j � r� t j | j | j d d �} t j | | j � } | j t j | j � | 7<| j d d � | j f | 8<t S)z� Add the value found in Step 4 to every element of each covered row, and subtract it from every element of each uncovered column. Return to Step 4 without altering any stars, primes, or covered lines. r2 r N)r r9 r r r3 r Z logical_notr; )r/ Zminvalr r r r< s $ r< )r, Znumpyr �fixesr r �objectr r r- r6 r; r= r< r r r r �<module> s ,0 ( % 2
Submit
FILE
FOLDER
Name
Size
Permission
Action
__init__.cpython-35.opt-1.pyc
13339 bytes
0644
__init__.cpython-35.pyc
13339 bytes
0644
_scipy_sparse_lsqr_backport.cpython-35.opt-1.pyc
14446 bytes
0644
_scipy_sparse_lsqr_backport.cpython-35.pyc
14446 bytes
0644
arpack.cpython-35.opt-1.pyc
51813 bytes
0644
arpack.cpython-35.pyc
51813 bytes
0644
bench.cpython-35.opt-1.pyc
484 bytes
0644
bench.cpython-35.pyc
484 bytes
0644
class_weight.cpython-35.opt-1.pyc
5471 bytes
0644
class_weight.cpython-35.pyc
5471 bytes
0644
deprecation.cpython-35.opt-1.pyc
2702 bytes
0644
deprecation.cpython-35.pyc
2702 bytes
0644
estimator_checks.cpython-35.opt-1.pyc
41678 bytes
0644
estimator_checks.cpython-35.pyc
41678 bytes
0644
extmath.cpython-35.opt-1.pyc
25515 bytes
0644
extmath.cpython-35.pyc
25515 bytes
0644
fixes.cpython-35.opt-1.pyc
11563 bytes
0644
fixes.cpython-35.pyc
11563 bytes
0644
graph.cpython-35.opt-1.pyc
4973 bytes
0644
graph.cpython-35.pyc
4973 bytes
0644
linear_assignment_.cpython-35.opt-1.pyc
7716 bytes
0644
linear_assignment_.cpython-35.pyc
7716 bytes
0644
metaestimators.cpython-35.opt-1.pyc
3925 bytes
0644
metaestimators.cpython-35.pyc
3925 bytes
0644
mocking.cpython-35.opt-1.pyc
2999 bytes
0644
mocking.cpython-35.pyc
2999 bytes
0644
multiclass.cpython-35.opt-1.pyc
12690 bytes
0644
multiclass.cpython-35.pyc
12690 bytes
0644
optimize.cpython-35.opt-1.pyc
4796 bytes
0644
optimize.cpython-35.pyc
4796 bytes
0644
random.cpython-35.opt-1.pyc
7804 bytes
0644
random.cpython-35.pyc
7804 bytes
0644
setup.cpython-35.opt-1.pyc
1923 bytes
0644
setup.cpython-35.pyc
1923 bytes
0644
sparsefuncs.cpython-35.opt-1.pyc
12432 bytes
0644
sparsefuncs.cpython-35.pyc
12546 bytes
0644
stats.cpython-35.opt-1.pyc
1797 bytes
0644
stats.cpython-35.pyc
1797 bytes
0644
testing.cpython-35.opt-1.pyc
27226 bytes
0644
testing.cpython-35.pyc
27330 bytes
0644
validation.cpython-35.opt-1.pyc
23261 bytes
0644
validation.cpython-35.pyc
23261 bytes
0644
N4ST4R_ID | Naxtarrr
