CODEWITHSUNDEEP
rggplot2gganimate
Matheus Barreto Alves
Matheus Barreto Alves

Reputation: 77

Making animations using gganimate: im not getting it right

I'm trying to make an COVID animation using the COVID data from my country. But i keep getting it wrong, and most of the issues i have no idea of how can i solve the problem.

libraries:
library(ggplot2)
library(tidyverse)
library(dplyr)
library(hrbrthemes)
library(rgdal)
library(raster)
library(ggmap)
library(tmap)
require(sp)
library(geobr)
library(readr)
library(gganimate)
library(gifski)

First of all, you can get the dataframe from here:

caso <- readr::read_csv("https://data.brasil.io/dataset/covid19/caso.csv.gz")
caso$date <- as.Date(caso$date)
caso$state <- as.factor(caso$state)
tibble [399,497 x 12] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
 $ date                          : Date[1:399497], format: "2020-07-22" "2020-07-21" "2020-07-20" ...
 $ state                         : Factor w/ 27 levels "AC","AL","AM",..: 4 4 4 4 4 4 4 4 4 4 ...
 $ city                          : chr [1:399497] NA NA NA NA ...
 $ place_type                    : chr [1:399497] "state" "state" "state" "state" ...
 $ confirmed                     : num [1:399497] 34660 34405 34145 33705 33585 ...
 $ deaths                        : num [1:399497] 544 533 515 507 505 499 493 488 483 478 ...
 $ order_for_place               : num [1:399497] 124 123 122 121 120 119 118 117 116 115 ...
 $ is_last                       : logi [1:399497] TRUE FALSE FALSE FALSE FALSE FALSE ...
 $ estimated_population_2019     : num [1:399497] 845731 845731 845731 845731 845731 ...
 $ city_ibge_code                : num [1:399497] 16 16 16 16 16 16 16 16 16 16 ...
 $ confirmed_per_100k_inhabitants: num [1:399497] 4098 4068 4037 3985 3971 ...
 $ death_rate                    : num [1:399497] 0.0157 0.0155 0.0151 0.015 0.015 0.0149 0.0149 0.01





> head(caso)
# A tibble: 6 x 12
  date       state city  place_type confirmed deaths order_for_place is_last estimated_popul~
  <date>     <fct> <chr> <chr>          <dbl>  <dbl>           <dbl> <lgl>              <dbl>
1 2020-07-22 AP    NA    state          34660    544             124 TRUE              845731
2 2020-07-21 AP    NA    state          34405    533             123 FALSE             845731
3 2020-07-20 AP    NA    state          34145    515             122 FALSE             845731
4 2020-07-19 AP    NA    state          33705    507             121 FALSE             845731
5 2020-07-18 AP    NA    state          33585    505             120 FALSE             845731
6 2020-07-17 AP    NA    state          33436    499             119 FALSE             845731
# ... with 3 more variables: city_ibge_code <dbl>, confirmed_per_100k_inhabitants <dbl>

The brazil map is also available:

Estados <- read_state(year=2018)

So far, i've been doing plots by summarizing the data,like this:

ggplot() +
  geom_sf(data=ontem, aes(fill=deaths), color="#FEBF57", size=.15, show.legend = TRUE) +
  labs(title = "Mortes por COVID",size=8) +
  scale_fill_distiller(palette = "BrBG",
                       name= "Mortes Confirmadas", limits=c(min(ontem$deaths),max(ontem$deaths)))+
  theme_void() + theme(plot.title = element_text(hjust = 0.5))

options(scipen=10000)

which results in this map: Covid Deaths - Brazil

Where "ontem" df is a dataframe of the last day status of the covid (subset of caso):

ontem <- caso %>% filter(date == Sys.Date()-1,place_type == 'state')

But i would like to make an animation of how the deaths (for example) increase each day, i tried to use something like the same code plus transition_time(date) but i keep getting warning/error messages.

Can someone help me with this? I'm stuck for days!

Upvotes: 0

Views: 228

Answers (2)

Adam Mar&#237;n
Adam Mar&#237;n

Reputation: 21

Ben is right, but the beauty of transition_time() is to use {frame_time} to see the specific time, usually I do something like this that I learned from someone else:

  scale_x_time(name = "Time", labels = label_time(format = "%H"), 
  breaks = as_hms(c("00:00:00","01:00:00","02:00:00","03:00:00",
  "04:00:00","05:00:00","06:00:00","07:00:00","08:00:00","09:00:00",
  "10:00:00","11:00:00","12:00:00","13:00:00","14:00:00","15:00:00",
  "16:00:00","17:00:00","18:00:00","19:00:00","20:00:00", 
  "21:00:00","22:00:00","23:00:00","24:00:00")))

Upvotes: 1

Ben
Ben

Reputation: 1154

The transition_time() function requires a vector to be in a date or time format. So, you must either ensure that your time variable is in a format that gganimate likes (it is pretty finicky with date formats) OR you could compute an integer that tracks sequence of time (1, 2, 3, 4...) after sorting by date/time, and using transition_states() with the sequence vector. The latter approach, I've found, is a lot easier.

Upvotes: 1

Related Questions