Is there an r function to produce excel 3d plot?

Question

Is there any way to produce excel 3d plot in r?

I have a matrix and I would like to plot it in 3d surface format using R. I managed to do it in excel and it gives me this plot https://drive.google.com/open?id=1hNqCguPX_ar8ueSK-u-yHTUYZxdyT_cE

I would like to produce same plot but with R this plot is produced with this data

m<-matrix(c(7,73,195,195,416,625,120,52,178,178,349,454,6,83,164,164,244,0,3,52,150,150,329,330),nrow = 4, ncol = 6, byrow = T)

maybe this is something you could do with the ``ggplot2`` and ``gg3D`` packages. — Gainz, May 31 '19 at 15:22
How to interpret this matrix? Are the columns / rows variable? Does each cell represent intensity? — bbiasi, May 31 '19 at 22:06
@Gainz Could you please provide me a code for this? I tried all possibilities but I failed at the end — Alaa, Jun 01 '19 at 16:06
@bbiasi Yes rows and columns are variables let say lat and long — Alaa, Jun 01 '19 at 16:08

bbiasi · Accepted Answer · 2019-06-03T16:55:19.850

Unfortunately the dataset in use is not available, so there is no way to faithfully reproduce the image and expand the details.

To create the contour surface, it is interesting to use the ggplot2 package with the geom_raster function. But this will return a continuous surface.

library(ggplot2)
library(dplyr)

x <- seq(1, 6, 1)
y <- seq(0, 3, 1)
grid <- expand.grid(x = x,
                    y = y)
m <- matrix(c(7,73,195,195,416,625,120,52,178,178,349,454,6,
            83,164,164,244,0,3,52,150,150,329,330), 
            nrow = 4, ncol = 6, byrow = T)
m2 <- m %>% 
  t %>% as.data.frame()
grid <- grid %>% 
  dplyr::mutate(response = c(m2$V1, m2$V2, m2$V3, m2$V4))

Palet <- c("royalblue2", "orangered3", "lavenderblush3", "gold3")

ggplot2::ggplot(grid, aes(x, y, z = response)) +
  geom_raster(aes(fill = response)) + 
  scale_fill_gradientn(colours = Palet, limits = c(0, 800)) +
  theme_void() +
  theme(legend.position = "bottom")

In addition, it is possible to perform data manipulation and work with the discrete fill. Where for each interval, a color will be assigned.

grid2 <- grid %>% 
  dplyr::mutate(cor = ifelse(response >= 0 & response < 200, 1,
                             ifelse(response >= 200 & response < 400, 2,
                                    ifelse(response >= 400 & response < 600, 3,
                                           ifelse(response >= 600, 4, "error")))))

ggplot2::ggplot(grid2, aes(x, y, z = response)) +
  geom_raster(aes(fill = cor)) + 
  scale_fill_manual(values = Palet, labels = c("0-200", "200-400",
                                               "400-600", "600-800"),
                    name = "") +
  theme_void() +
  theme(legend.position = "bottom")

However, if it is considered that in matrix m each cell is intensity, it is possible to generate a 3D plot.

library(plot3D)
persp3D(z = m, theta = 60)

3D surface plot from 2D matrix

Plot 3D data in R

Impressive package for 3D and 4D graph - R software and data visualization

Thanks for help, however it is still very far from plot made by Microsoft excel (attached) — Alaa, Jun 01 '19 at 16:09
@Alaa see this: [link1](https://meta.stackexchange.com/questions/5234/how-does-accepting-an-answer-work), [link2](https://meta.stackexchange.com/questions/173399/how-to-upvote-on-stack-overflow). Thanks. — bbiasi, Jun 12 '19 at 13:25

Is there an r function to produce excel 3d plot?

1 Answers1