Remove white line or space from between geom_tiles

John_Ryan_Kivela · December 26, 2022, 8:20pm

I am trying to make a heatmap using geom_tile. It is looking pretty good, but there are either white lines, or spaces between certain columns of tiles. The plot is not doing it for all of the columns. Any advice would be helpful. I included the code chunk below and a glimpse of the data. This is my first post, so I apologize if I haven't included code and data properly.

VBP_Data2 %>% 
  ggplot(aes(Date,
             Provider)) +
  
  geom_tile(aes(fill = PercentCompliant))+
 
  facet_grid(cols = vars(SubMeasureID),
             scales = "fixed",
             space = "fixed")+
  
  scale_fill_gradientn(colours = c("#9a0138","#f688ee","#000775"), 
                       values = rescale(c(0,.5,1.0)),
                       guide = "colorbar",
                       limits = c(0,1.0))+
  
  theme_minimal()+

  theme(axis.text.x = element_text(
    angle = 45, 
    vjust = 1.1, 
    hjust=1,
    size = 14),
    strip.text = element_text(
      size = 14),
    axis.text.y = element_text(
      size = 14),
    plot.title = element_text(
      size = 16),
    axis.title.x.bottom = element_text(
      size = 16)
  )+

  ylab(NULL)+
  xlab ("Alliance Provider Baseline Performance")

> glimpse(VBP_Data2)
Rows: 189
Columns: 7
$ Date             <date> 2022-04-01, 2022-04-01, 2022-04-01, 2022-04-01, 2022-0…
$ Provider         <chr> "A", "B", "…
$ SubMeasureID     <chr> "AMM2", "AMM2", "AMM2", "AMM2", "AMM2", "AMM2", "AMM2",…
$ Compliant        <dbl> 36, 43, 68, 77, 85, 96, 139, 238, 290, 2, 1, 2, 1, 1, 3…
$ NonCompliant     <dbl> 25, 33, 54, 58, 63, 67, 99, 169, 216, 1, 3, 4, 9, 11, 1…
$ TotalEligible    <dbl> 61, 76, 122, 135, 148, 163, 238, 407, 506, 3, 4, 6, 10,…
$ PercentCompliant <dbl> 0.59016393, 0.56578947, 0.55737705, 0.57037037,

jrkrideau · December 26, 2022, 8:28pm

Hi welcome aboard.

The code looks fine but the data is a bit dicey. A very handy way to supply some sample data is the dput() function. In the case of a large dataset something like dput(head(mydata, 100)) should supply the data we need.

Just do dput(mydata) where mydata is your data. Copy the output and paste it here.

John_Ryan_Kivela · December 26, 2022, 8:50pm

Thank you for you response, and the tip!

How is this:

>  dput(head(VBP_Data2, 100))
structure(list(Date = structure(c(19083, 19083, 19083, 19083, 
19083, 19083, 19083, 19083, 19083, 19083, 19083, 19083, 19083, 
19083, 19083, 19083, 19083, 19083, 19083, 19083, 19083, 19083, 
19083, 19083, 19083, 19083, 19083, 19113, 19113, 19113, 19113, 
19113, 19113, 19113, 19113, 19113, 19113, 19113, 19113, 19113, 
19113, 19113, 19113, 19113, 19113, 19113, 19113, 19113, 19113, 
19113, 19113, 19113, 19113, 19113, 19144, 19144, 19144, 19144, 
19144, 19144, 19144, 19144, 19144, 19144, 19144, 19144, 19144, 
19144, 19144, 19144, 19144, 19144, 19144, 19144, 19144, 19144, 
19144, 19144, 19144, 19144, 19144, 19174, 19174, 19174, 19174, 
19174, 19174, 19174, 19174, 19174, 19174, 19174, 19174, 19174, 
19174, 19174, 19174, 19174, 19174, 19174), class = "Date"), Provider = c("Encompass", 
"Little Colorado Behavioral Health", "Spectrum", "The Guidance Center", 
"Community Bridges", "Polara", "Changepoint", "Mohave Mental Health", 
"Southwest Behavioral and Health", "Encompass", "Little Colorado Behavioral Health", 
"The Guidance Center", "Community Bridges", "Polara", "Spectrum", 
"Changepoint", "Mohave Mental Health", "Southwest Behavioral and Health", 
"Encompass", "Community Bridges", "The Guidance Center", "Spectrum", 
"Little Colorado Behavioral Health", "Polara", "Changepoint", 
"Southwest Behavioral and Health", "Mohave Mental Health", "Encompass", 
"Little Colorado Behavioral Health", "Spectrum", "The Guidance Center", 
"Community Bridges", "Polara", "Changepoint", "Mohave Mental Health", 
"Southwest Behavioral and Health", "Encompass", "Little Colorado Behavioral Health", 
"Community Bridges", "The Guidance Center", "Polara", "Spectrum", 
"Changepoint", "Mohave Mental Health", "Southwest Behavioral and Health", 
"Encompass", "The Guidance Center", "Community Bridges", "Spectrum", 
"Little Colorado Behavioral Health", "Polara", "Changepoint", 
"Southwest Behavioral and Health", "Mohave Mental Health", "Encompass", 
"Little Colorado Behavioral Health", "Spectrum", "The Guidance Center", 
"Community Bridges", "Polara", "Changepoint", "Mohave Mental Health", 
"Southwest Behavioral and Health", "Encompass", "Little Colorado Behavioral Health", 
"Community Bridges", "The Guidance Center", "Spectrum", "Polara", 
"Changepoint", "Mohave Mental Health", "Southwest Behavioral and Health", 
"Encompass", "The Guidance Center", "Spectrum", "Little Colorado Behavioral Health", 
"Polara", "Changepoint", "Southwest Behavioral and Health", "Mohave Mental Health", 
"Community Bridges", "Encompass", "Little Colorado Behavioral Health", 
"Spectrum", "The Guidance Center", "Community Bridges", "Polara", 
"Changepoint", "Mohave Mental Health", "Southwest Behavioral and Health", 
"Encompass", "Little Colorado Behavioral Health", "Community Bridges", 
"The Guidance Center", "Spectrum", "Polara", "Changepoint", "Mohave Mental Health", 
"Southwest Behavioral and Health", "Encompass"), SubMeasureID = c("AMM2", 
"AMM2", "AMM2", "AMM2", "AMM2", "AMM2", "AMM2", "AMM2", "AMM2", 
"FUH7", "FUH7", "FUH7", "FUH7", "FUH7", "FUH7", "FUH7", "FUH7", 
"FUH7", "HDO", "HDO", "HDO", "HDO", "HDO", "HDO", "HDO", "HDO", 
"HDO", "AMM2", "AMM2", "AMM2", "AMM2", "AMM2", "AMM2", "AMM2", 
"AMM2", "AMM2", "FUH7", "FUH7", "FUH7", "FUH7", "FUH7", "FUH7", 
"FUH7", "FUH7", "FUH7", "HDO", "HDO", "HDO", "HDO", "HDO", "HDO", 
"HDO", "HDO", "HDO", "AMM2", "AMM2", "AMM2", "AMM2", "AMM2", 
"AMM2", "AMM2", "AMM2", "AMM2", "FUH7", "FUH7", "FUH7", "FUH7", 
"FUH7", "FUH7", "FUH7", "FUH7", "FUH7", "HDO", "HDO", "HDO", 
"HDO", "HDO", "HDO", "HDO", "HDO", "HDO", "AMM2", "AMM2", "AMM2", 
"AMM2", "AMM2", "AMM2", "AMM2", "AMM2", "AMM2", "FUH7", "FUH7", 
"FUH7", "FUH7", "FUH7", "FUH7", "FUH7", "FUH7", "FUH7", "HDO"
), Compliant = c(36, 43, 68, 77, 85, 96, 139, 238, 290, 2, 1, 
2, 1, 1, 3, 2, 12, 15, 2, 1, 2, 6, 6, 6, 38, 28, 17, 39, 55, 
86, 94, 105, 119, 196, 286, 355, 7, 5, 6, 13, 20, 24, 21, 67, 
49, 4, 5, 13, 12, 20, 13, 70, 55, 39, 41, 57, 89, 95, 104, 127, 
206, 288, 367, 9, 6, 9, 19, 34, 26, 26, 81, 62, 4, 6, 12, 20, 
12, 63, 55, 38, 10, 43, 60, 93, 98, 107, 136, 215, 296, 381, 
8, 11, 11, 26, 38, 43, 47, 87, 94, 4), NonCompliant = c(25, 33, 
54, 58, 63, 67, 99, 169, 216, 1, 3, 4, 9, 11, 10, 25, 15, 26, 
22, 33, 33, 29, 53, 62, 117, 193, 223, 35, 40, 72, 75, 74, 88, 
122, 188, 247, 3, 7, 22, 34, 28, 24, 42, 45, 71, 82, 105, 103, 
120, 148, 217, 415, 686, 868, 34, 42, 74, 80, 76, 99, 124, 197, 
261, 5, 12, 30, 38, 27, 38, 56, 70, 101, 81, 110, 120, 145, 216, 
405, 702, 868, 105, 30, 38, 68, 72, 73, 89, 115, 188, 252, 11, 
16, 35, 43, 40, 55, 80, 97, 131, 87), TotalEligible = c(61, 76, 
122, 135, 148, 163, 238, 407, 506, 3, 4, 6, 10, 12, 13, 27, 27, 
41, 24, 34, 35, 35, 59, 68, 155, 221, 240, 74, 95, 158, 169, 
179, 207, 318, 474, 602, 10, 12, 28, 47, 48, 48, 63, 112, 120, 
86, 110, 116, 132, 168, 230, 485, 741, 907, 75, 99, 163, 175, 
180, 226, 330, 485, 628, 14, 18, 39, 57, 61, 64, 82, 151, 163, 
85, 116, 132, 165, 228, 468, 757, 906, 115, 73, 98, 161, 170, 
180, 225, 330, 484, 633, 19, 27, 46, 69, 78, 98, 127, 184, 225, 
91), PercentCompliant = c(0.590163934426229, 0.565789473684211, 
0.557377049180328, 0.57037037037037, 0.574324324324324, 0.588957055214724, 
0.584033613445378, 0.584766584766585, 0.573122529644269, 0.666666666666667, 
0.25, 0.333333333333333, 0.1, 0.0833333333333333, 0.230769230769231, 
0.0740740740740741, 0.444444444444444, 0.365853658536585, 0.916666666666667, 
0.970588235294118, 0.942857142857143, 0.828571428571429, 0.898305084745763, 
0.911764705882353, 0.754838709677419, 0.873303167420814, 0.929166666666667, 
0.527027027027027, 0.578947368421053, 0.544303797468354, 0.556213017751479, 
0.58659217877095, 0.57487922705314, 0.616352201257862, 0.60337552742616, 
0.589700996677741, 0.7, 0.416666666666667, 0.214285714285714, 
0.276595744680851, 0.416666666666667, 0.5, 0.333333333333333, 
0.598214285714286, 0.408333333333333, 0.953488372093023, 0.954545454545455, 
0.887931034482759, 0.909090909090909, 0.880952380952381, 0.943478260869565, 
0.855670103092784, 0.925775978407557, 0.957001102535832, 0.546666666666667, 
0.575757575757576, 0.54601226993865, 0.542857142857143, 0.577777777777778, 
0.561946902654867, 0.624242424242424, 0.593814432989691, 0.584394904458599, 
0.642857142857143, 0.333333333333333, 0.230769230769231, 0.333333333333333, 
0.557377049180328, 0.40625, 0.317073170731707, 0.536423841059603, 
0.380368098159509, 0.952941176470588, 0.948275862068966, 0.909090909090909, 
0.878787878787879, 0.947368421052632, 0.865384615384615, 0.927344782034346, 
0.958057395143488, 0.91304347826087, 0.589041095890411, 0.612244897959184, 
0.577639751552795, 0.576470588235294, 0.594444444444444, 0.604444444444444, 
0.651515151515151, 0.611570247933884, 0.601895734597156, 0.421052631578947, 
0.407407407407407, 0.239130434782609, 0.376811594202899, 0.487179487179487, 
0.438775510204082, 0.37007874015748, 0.472826086956522, 0.417777777777778, 
0.956043956043956)), row.names = c(NA, -100L), class = c("tbl_df", 
"tbl", "data.frame"))

technocrat · December 26, 2022, 9:22pm

Here's where it's happening—haven't figured out what to do, though.

John_Ryan_Kivela · December 26, 2022, 9:28pm

I was pondering if it has to do with the number of days in the month. Thirty days has September, and so forth. The problematic months seem to be May, July, and August, which have 31...

technocrat · December 26, 2022, 9:49pm

Ah, got an idea for fix

technocrat · December 26, 2022, 10:12pm

Thought I could use this

library(lubridate)
jigger <- function(x) paste(month(x),year(x))
VBP_Data2 %>% 
  ggplot(aes(Date),
             Provider) +
  x-axis = jigger

but might need different way of setting labels, since this chokes.

John_Ryan_Kivela · December 27, 2022, 1:09am

Thanks for the reply. I'm afraid I don't understand how to use this one.

technocrat · December 27, 2022, 1:50am

Of course. Because I botched it.


ggplot(aes(jiggerDate)),

But it doesn’t work anyway. Will return later 2nite PST

scottyd22 · December 27, 2022, 3:31am

I'm not sure why the white space is there to begin with, but if you change the Date in your plot to character, the white space goes away.

VBP_Data2 %>% 
  ggplot(aes(as.character(Date),
             Provider)) +
  ...

John_Ryan_Kivela · December 29, 2022, 6:34pm

Thanks! This does the trick.

I liked the look of having the mmm-yy format on the x-axis, but beggars can't be choosers.

New question: Can you help me get the scale to display as a percent of 100%? I mutate the variable to a percent, and it shows as a percent in glimpse, but shows as a number between 1 and 0 in the guide on the plot.

scottyd22 · December 29, 2022, 7:35pm

Add labels = scales::percent to this section.

scale_fill_gradientn(colours = c("#9a0138","#f688ee","#000775"), 
                       values = scales::rescale(c(0,.5,1.0)),
                       guide = "colorbar",
                       limits = c(0,1.0),
                       labels = scales::percent)+

John_Ryan_Kivela · December 29, 2022, 7:41pm

Hazah!

Thanks for the assist(s). I learned a lot today!

Thank you.

system · January 19, 2023, 7:41pm

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.