我在R中尝试这个代码并且闪亮。我无法执行HeatMap，图表没有显示。它在R控制台中显示以下错误 -
eval中的错误（替换（expr），envir，enclos）： 设定美学的长度不相容：颜色，大小 错误：设置美学的长度不兼容：颜色，尺寸
任何人都可以告诉我出了什么问题吗？
代码：
shinyServer(function(input, output) {
## Testing values
if (FALSE) {
input <- list(poi = "London Eye",
start = "2013-01-01",
months = 6,
facet = "none",
type = "roadmap",
res = TRUE,
bw = FALSE,
zoom = 14,
alpharange = c(0.1, 0.4),
bins = 15,
boundwidth = 0.1,
boundcolour = "grey95",
low = "yellow",
high = "red",
watermark = "TRUE")
}
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
## Static Functions
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
## Get data from API (using jsonlite to flatten list quickly)
get_data <- function(map.geocode, map.period) {
## Loop through all months
for (n_period in 1:length(map.period)) {
## Create an URL for API
url_api <- paste0("http://data.police.uk/api/crimes-street/all-crime?",
"lat=", map.geocode[2],
"&lng=", map.geocode[1],
"&date=", map.period[n_period])
## Download data
tmp_dat <- jsonlite::fromJSON(getURL(url_api), flatten = TRUE)
## Store data
if (n_period == 1) out_dat <- tmp_dat else out_dat <- rbind(out_dat, tmp_dat)
}
## Trim and Rename columns
col_inc <- c("category", "location_type", "id", "month", "location.latitude",
"location.longitude", "location.streed.id",
"location.street.name")
out_dat <- out_dat[, c(1,2,5,7,8,9,10,11)]
colnames(out_dat) <- c("category", "type", "id", "month", "latitude",
"longitude", "street_id", "street_name")
## Convert characters into numerical values
out_dat$id <- as.integer(out_dat$id)
out_dat$latitude <- as.numeric(out_dat$latitude)
out_dat$longitude <- as.numeric(out_dat$longitude)
out_dat$street_id <- as.integer(out_dat$street_id)
## Return
return(out_dat)
}
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
## Reactive Functions
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
## Get Geocode
map.geocode <- reactive({
suppressMessages(data.frame(geocode = geocode(paste(input$poi, "UK"))))
})
## Define Period
map.period <- reactive({
format(seq(as.Date(input$start), length=input$months, by="months"), "%Y-%m")
})
## Create Data Frame
create.df <- reactive({
## Use Reactive Functions to download data and create a data frame
temp.geocode <- map.geocode()
temp.period <- map.period()
df <- get_data(temp.geocode, temp.period)
## Return
df
})
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
## Output 1 - Data Table
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
output$datatable <- renderDataTable({
## Check input
if (input$poi == "London Eye (Demo)") {
## Use preloaded data
load("./demo/demo_london_eye.rda")
} else {
## Use reactive function
df <- create.df()
}
## Display df
df
}, options = list(aLengthMenu = c(10, 15, 30, 50, 100, 500), iDisplayLength = 15))
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
## Output 2 - Heat Map
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
output$map <- renderPlot({
## Check input
if (input$poi == "London Eye (Demo)") {
## Use preloaded data
load("./demo/demo_london_eye.rda")
} else {
## Use reactive functions
temp.geocode <- map.geocode()
temp.period <- map.period()
## Get df
df <- create.df()
## Create a data frame for the map center (point of interest)
center.df <- data.frame(temp.geocode, location = input$poi)
colnames(center.df) <- c("lon","lat","location")
}
## Download base map using {ggmap}
## Note that a PNG file "ggmapTemp.png" will be created
## The PNG is not needed for the analysis, you can delete it later
## Define colour and scale of base map
if (input$bw) temp.color <- "bw" else temp.color <- "color"
if (input$res) temp.scale <- 2 else temp.scale <- 1
map.base <- get_googlemap(
as.matrix(temp.geocode),
maptype = input$type, ## Map type as defined above (roadmap, terrain, satellite, hybrid)
markers = temp.geocode,
zoom = input$zoom, ## 14 is just about right for a 1-mile radius
color = temp.color, ## "color" or "bw" (black & white)
scale = temp.scale ## Set it to 2 for high resolution output
)
## Convert the base map into a ggplot object
## All added Cartesian coordinates to enable more geom options later on
map.base <- ggmap(map.base, extend = "panel") + coord_cartesian() + coord_fixed(ratio = 1.5)
## Main ggplot
map.final <- map.base +
## Create a density plot
## based on the ggmap's crime data example
stat_density2d(aes(x = longitude,
y = latitude,
fill = ..level..,
alpha = ..level..),
na.rm = TRUE,
size = input$boundwidth,
bins = input$bins, ## Change and experiment with no. of bins
data = df,
geom = "polygon",
colour = input$boundcolour) +
## Configure the scale and panel
scale_fill_gradient(low = input$low, high = input$high) +
scale_alpha(range = input$alpharange) +
## Title and labels
labs(x = "Longitude", y = "Latitude") +
ggtitle(paste("Crimes in/around ", center.df$location,
" from ", temp.period[1],
" to ", temp.period[length(temp.period)], sep="")) +
## Other theme settings
theme_bw() +
theme(
plot.title = element_text(size = 26, face = 'bold', vjust = 2),
axis.text = element_blank(),
axis.title = element_blank(),
axis.ticks = element_blank(),
legend.position = "none",
strip.background = element_rect(fill = 'grey80'),
strip.text = element_text(size = 18)
)
# Use Watermark?
if (input$watermark) {
if (input$facet == "none") {
map.final <- map.final + annotate("text", x = center.df$lon, y = -Inf,
label = "www.jofaichow.co.uk",
vjust = -1.5, col = "steelblue",
cex = 10,
fontface = "bold", alpha = 0.5)
}
}
## Use Facet?
if (input$facet == "type") {map.final <- map.final + facet_wrap(~ type)}
if (input$facet == "month") {map.final <- map.final + facet_wrap(~ month)}
if (input$facet == "category") {map.final <- map.final + facet_wrap(~ category)}
## Display ggplot2 object
print(map.final)
}, width = 900, height = 900)
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
## Output 3 - Trends 1
## ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
output$trends1 <- renderPlot({
## Check input
if (input$poi == "London Eye (Demo)") {
## Use preloaded data
load("./demo/demo_london_eye.rda")
} else {
## Use Reactive Functions
temp.geocode <- map.geocode()
temp.period <- map.period()
## Get df
df <- create.df()
## Create a data frame for the map center (point of interest)
center.df <- data.frame(temp.geocode, location = input$poi)
colnames(center.df) <- c("lon","lat","location")
}
## Create bar chart summary
plot1 <- ggplot(df, aes(x = month, fill = category)) +
geom_bar(colour = "black") + facet_wrap(~ category) +
labs(x = "Months", y = "Crime Records") +
ggtitle(paste("Crimes in/around ", center.df$location,
": Trends from ", temp.period[1],
" to ", temp.period[length(temp.period)], sep="")) +
theme_bw() +
theme(
plot.title = element_text(size = 26, face = 'bold', vjust = 2),
axis.text.x = element_blank(),
axis.text = element_text(size = 16),
axis.title.x = element_text(size = 24),
axis.title.y = element_text(size = 24),
axis.ticks.x = element_blank(),
strip.background = element_rect(fill = 'grey80'),
strip.text.x = element_text(size = 18),
legend.position = "none",
panel.grid = element_blank()
)
## Print
print(plot1)
}, width = 1000, height = 900)
})