如何在Prolog中围绕其中心点旋转4 x 4矩阵?我可以简单地在4 x 4矩阵的情况下重新排列元素,但是对于像N x N这样的一般情况,如何做到这一点?
答案 0 :(得分:3)
你想要的是不完全矩阵换位......但差不多!
:- use_module(library(clpfd)). matrix_rotated(Xss, Zss) :- transpose(Xss, Yss), maplist(reverse, Yss, Zss).
示例查询:
?- matrix_rotated([[a1,a2,a3,a4], [b1,b2,b3,b4], [c1,c2,c3,c4]], Xss). Xss = [[c1,b1,a1], [c2,b2,a2], [c3,b3,a3], [c4,b4,a4]].
答案 1 :(得分:3)
有人是wrong on the internet,所以值班。 毕竟,我们必须遵守Cunningham's Law。
此答案充分利用了library(clpfd)中的SWI-Prolog。
:- use_module(library(clpfd)).
rotate_clockwise(Matrix, N, Rotated) :-
N_mod_4 #= N mod 4,
rotate_clockwise_(N_mod_4, Matrix, Rotated).
rotate_clockwise_(0, M, M).
rotate_clockwise_(1, M, R) :-
transpose(M, R0),
maplist(reverse, R0, R).
rotate_clockwise_(2, M, R) :-
reverse(M, R0),
maplist(reverse, R0, R).
rotate_clockwise_(3, M, R) :-
transpose(M, R0),
reverse(R0, R).
您可以通过将矩形矩阵翻转两次来旋转它(用书或其他东西试试)。 根据您翻转的两个轴,您可以获得3个非平凡旋转中的一个:如果我们同意"一次旋转"顺时针旋转90度,然后你可以旋转0,1,2或3次。在上面的代码中:
reverse
maplist(reverse)
transpose
至于为什么不定义旋转90度顺时针旋转然后旋转多次:好吧,它实际上是更多的代码!
根据上面的定义,无论我们给rotate_clockwise/3
提供什么有效参数,它都会进行正确的轮换,不进行不必要的翻转,并使调用者代码保持简短。
?- rotate_clockwise([[a,b,c],[d,e,f]], -1, R). % once counter-clockwise
R = [[c, f], [b, e], [a, d]].
?- rotate_clockwise([[a,b,c],[d,e,f]], 2, R). % 180 degrees
R = [[f, e, d], [c, b, a]].
?- rotate_clockwise([[a,b,c],[d,e,f]], 168, R). % rotate once each hour, for a week
R = [[a, b, c], [d, e, f]].
?- rotate_clockwise([[a,b,c],[d,e,f]], N, R). % enumerate all possibilities
R = [[a, b, c], [d, e, f]],
N mod 4#=0 ;
R = [[d, a], [e, b], [f, c]],
N mod 4#=1 ;
R = [[f, e, d], [c, b, a]],
N mod 4#=2 ;
R = [[c, f], [b, e], [a, d]],
N mod 4#=3.
?- Matrix = [[a],[b]],
rotate_clockwise(Matrix, -1, R),
rotate_clockwise(Matrix, 3, R). % those two mean the same
Matrix = [[a], [b]],
R = [[a, b]].
答案 2 :(得分:1)
如上所述,transpose / 2不是正确的解决方案。我编写了一些东西(可能有点太一般),允许指定转移到右边的单元格数量的旋转量。 Shift默认为1,但后来我注意到为了获得90°旋转,换档在传递到内框时减小。将匿名变量传递给matrix_rotate / 3具有使用当前内部帧大小的效果。然后,它顺时针旋转90°。
/** <module> matrix_rotate
*
* answer to http://stackoverflow.com/questions/35594027/rotate-a-matrix-in-prolog
* --------
*
* source file /home/carlo/prolog/matrix_rotate.pl
* created at mer feb 24 16:43:50 2016
*
* @author carlo
* @version 0.9.9
* @copyright carlo
* @license LGPL v2.1
*/
:- module(matrix_rotate,
[matrix_rotate/2
,matrix_rotate/3
,matrix_allocate/4
,matrix_row_col_element/4
,rowcol/3
]).
:- meta_predicate matrix_allocate(3, +,+,-).
matrix_allocate(Pred, NRows, NCols, Mat) :-
bagof(Vs, Row^(
between(1, NRows, Row),
bagof(C, Col^(between(1, NCols, Col), call(Pred, Row, Col, C)), Vs)
), Mat).
empty(_R,_C,_).
rowcol(R, C, (R,C)).
matrix_rotate(Mat, Rot) :-
matrix_rotate(Mat, 1, Rot).
matrix_rotate(Mat, Shift, Rot) :-
matrix_is_square(Mat, Size),
matrix_allocate(empty, Size, Size, Rot),
frames_shift(Mat, Shift, 1, Rot),
( Size mod 2 =:= 1
-> Cen is Size // 2 + 1,
matrix_row_col_element(Mat, Cen, Cen, E),
matrix_row_col_element(Rot, Cen, Cen, E)
; true
).
frames_shift(Mat, Shift, Frame, Rot) :-
length(Mat, Size),
Frame =< Size // 2,
( integer(Shift)
-> ActualShift = Shift
; ActualShift is Size - (Frame - 1)*2 - 1
),
frame(Mat, Frame, L),
shift_right(L, ActualShift, S),
frame(Rot, Frame, S),
F is Frame+1,
!, frames_shift(Mat, Shift, F, Rot).
frames_shift(_Mat, _Shift, _Frame, _Rot).
matrix_is_square(Mat, Size) :-
length(Mat, Size),
forall(member(Row, Mat), length(Row, Size)).
matrix_row_col_element(Mat, Row, Col, El) :-
nth1(Row, Mat, Cells),
nth1(Col, Cells, El).
shift_right(List, Shift, Shifted) :-
length(Post, Shift),
append(Pre, Post, List),
append(Post, Pre, Shifted).
frame(Mat, N, Frame) :-
length(Mat, S),
T is N, B is S-N+1,
L is N, R is S-N+1,
matrix_range_elements(Mat, T,T, L,R-1, Top),
matrix_range_elements(Mat, T,B-1, R,R, Right),
matrix_range_elements(Mat, B,B, R,L+1, Bottom),
matrix_range_elements(Mat, B,T+1, L,L, Left),
append([Top, Right, Bottom, Left], Frame).
matrix_range_elements(Mat, RStart, RStop, CStart, CStop, Elems) :-
bagof(E, matrix_element_between(Mat, RStart, RStop, CStart, CStop, E), Elems).
matrix_element_between(Mat, RStart, RStop, CStart, CStop, Elem) :-
signed_between(RStart, RStop, R),
signed_between(CStart, CStop, C),
matrix_row_col_element(Mat, R, C, Elem).
signed_between(Start, Stop, I) :-
A is Start, B is Stop,
( B < A -> between(B, A, N), I is A+B-N ; between(A, B, I) ).
此片段突出显示了将矩阵作为列表列表处理时有用的一般模式。还有一些困难......
使用示例:
?- matrix_allocate(rowcol,4,4,M),matrix_rotate(M,_,R),maplist(writeln,R).
[(4,1),(3,1),(2,1),(1,1)]
[(4,2),(3,2),(2,2),(1,2)]
[(4,3),(3,3),(2,3),(1,3)]
[(4,4),(3,4),(2,4),(1,4)]
M = [[(1, 1), (1, 2), (1, 3), (1, 4)], [(2, 1), (2, 2), (2, 3), (2, 4)], [(3, 1), (3, 2), (3, 3), (3, 4)], [(4, 1), (4, 2), (4, 3), (4, 4)]],
R = [[(4, 1), (3, 1), (2, 1), (1, 1)], [(4, 2), (3, 2), (2, 2), (1, 2)], [(4, 3), (3, 3), (2, 3), (1, 3)], [(4, 4), (3, 4), (2, 4), (1, 4)]].
?- matrix_allocate(rowcol,3,3,M),matrix_rotate(M,_,R),maplist(writeln,R).
[(3,1),(2,1),(1,1)]
[(3,2),(2,2),(1,2)]
[(3,3),(2,3),(1,3)]
M = [[(1, 1), (1, 2), (1, 3)], [(2, 1), (2, 2), (2, 3)], [(3, 1), (3, 2), (3, 3)]],
R = [[(3, 1), (2, 1), (1, 1)], [(3, 2), (2, 2), (1, 2)], [(3, 3), (2, 3), (1, 3)]].