可视化和旋转矩阵 [英] Visualising and rotating a matrix

问题描述

我正在尝试可视化一个大约为500x500的上三角矩阵.除此之外，我还尝试旋转该图像，以使其看起来像三角形朝上:

I am trying to visualise an upper triangular matrix that is approximately 500x500. Alongside this I am also trying to rotate that image so that it looks like the triangle is pointing upward:

(这是通过拍摄图形设备的快照然后旋转该图像来实现的).

(This was achieved by taking a snapshot of a graphics device and then rotating that image).

在该图像中，每一列和每一行都需要指定宽度.

As in that image each column and row needs to have it's width specified.

我尝试将image()函数与grid包一起使用(使用旋转了45度的观察面板)，但是不起作用.有谁知道更好的解决方案?

I have tried using the image() function with the grid package (using a viewpanel that is rotated by 45 degrees) however that does not work. Does anybody know a better solution?

推荐答案

以下是使用基础图形的rasterImage的简单愚蠢方法:

Here a simple and stupid approach using base graphics' rasterImage:

plotTriMatrix <- function(x) {
  ## clear lower triangle
  x[lower.tri(x)] <- NA

  ## calculate diag
  nr <- nrow(x)
  nc <- ncol(x)
  d <- sqrt(nr^2 + nc^2)
  d2 <- 0.5 * d

  ## empty plot area
  plot(NA, type="n", xlim=c(0, d), ylim=c(0, d), xlab="", ylab="", asp=1)

  ## plot matrix and rotate 45
  rasterImage(as.raster(x),
              xleft=d2, xright=d2+nc, ybottom=-d2, ytop=-d2+nr,
              interpolate=FALSE, angle=45)
}

示例:

set.seed(123)
m <- matrix(runif(100), 10, 10)

plotTriMatrix(m)

这篇关于可视化和旋转矩阵的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！