如何用gganimate保存与原ggplot样式一致的高质量GIF？

你的代码运行正常，但保存GIF时出现的画质损失、文本错位和宽高比问题，主要源于宽高比不统一、固定文本误用数据绑定的geom_text以及GIF渲染参数不足这几个核心问题，以下是针对性的修复方案：

核心修复点与修改后的代码 #

1. 统一宽高比，避免变形 #

静态图采用12:7的比例，手动设置的960x560虽接近但易有误差，通过比例计算自动生成GIF尺寸，确保完全匹配：

# 基于静态图比例计算GIF尺寸，保证宽高比严格一致
gif_width <- 1200  # 更高的基础宽度直接提升画质
gif_height <- round(gif_width * (plot_height / plot_width))

2. 修复文本注释错位 #

你用geom_text(aes(...))添加固定位置的文本，但geom_text会随数据行重复渲染（比如uncount后的数据），导致文本错位或重叠。所有固定位置的文本应改用annotate()，脱离数据绑定：

# 替换所有固定位置的geom_text为annotate
annotate("text", x = -4, y = 0.0100, label = "Varones", fontface = "bold", color = "#6A6654", vjust = -.5, family = "Verdana") +
annotate("text", x = -4, y = -.03, label = "Mujeres", fontface = "bold", color = "#D90D0D", vjust = -.5, family = "Verdana") +
annotate("text", x = -.5, y = -.31, label = "-31%", fontface = "bold", color = "#D90D0D", vjust = -.5, family = "Verdana") +
annotate("text", x = 10, y = -.38, label = "-39%", fontface = "bold", color = "#D90D0D", vjust = -.5, family = "Verdana") +

3. 提升GIF画质与渲染质量 #

在anim_save中指定与静态图一致的dpi，并用magick_renderer控制输出质量，同时调整帧参数平衡流畅度与文件大小：

anim_save(
  "motherhood_penalties_uy.gif", 
  animation = anim, 
  end_pause = 10, 
  nframes = 200,  # 增加帧数提升动画流畅度
  fps = 20,       # 降低帧率避免过快，同时控制文件体积
  width = gif_width, 
  height = gif_height,
  dpi = dpi_setting,      # 和静态图一致的dpi，保证画质统一
  renderer = magick_renderer(quality = 100)  # 最高质量渲染
)

修改后的完整代码 #

pacman::p_load(gganimate, tidyverse, here)

# Setting consistent dimensions for both static and animated outputs
plot_width <- 12
plot_height <- 7
dpi_setting <- 300

# Motherhood Penalties | Uruguay

uruguay <- read_csv("name,confHigh,confLow,estimate,gender,t_es
                 Uruguay,0.0236,-7e-4,0.0114,male,-5
                 Uruguay,0.0276,-0.0589,-0.0157,female,-5
                 Uruguay,0.0087,-0.0157,-0.0035,male,-4
                 Uruguay,0.03,-0.048,-0.009,female,-4
                 Uruguay,0.0257,0.0028,0.0142,male,-3
                 Uruguay,0.0287,-0.0433,-0.0073,female,-3
                 Uruguay,0,0,0,male,-2
                 Uruguay,0,0,0,female,-2
                 Uruguay,0.0113,-0.0115,-1e-4,male,-1
                 Uruguay,0.0042,-0.0602,-0.028,female,-1
                 Uruguay,0.022,1e-4,0.011,male,0
                 Uruguay,-0.2773,-0.3384,-0.3079,female,0
                 Uruguay,0.0281,0.0064,0.0173,male,1
                 Uruguay,-0.2816,-0.3413,-0.3115,female,1
                 Uruguay,0.0188,-0.003,0.0079,male,2
                 Uruguay,-0.3002,-0.3575,-0.3288,female,2
                 Uruguay,0.0188,-0.0029,0.0079,male,3
                 Uruguay,-0.3197,-0.3748,-0.3472,female,3
                 Uruguay,0.0266,0.0052,0.0159,male,4
                 Uruguay,-0.3283,-0.3821,-0.3552,female,4
                 Uruguay,0.0192,-0.0027,0.0082,male,5
                 Uruguay,-0.3554,-0.4083,-0.3818,female,5
                 Uruguay,0.019,-0.0029,0.008,male,6
                 Uruguay,-0.3527,-0.4048,-0.3787,female,6
                 Uruguay,0.0195,-0.0025,0.0085,male,7
                 Uruguay,-0.3695,-0.421,-0.3953,female,7
                 Uruguay,0.0137,-0.0089,0.0024,male,8
                 Uruguay,-0.3565,-0.407,-0.3818,female,8
                 Uruguay,0.0163,-0.0062,0.0051,male,9
                 Uruguay,-0.3636,-0.4137,-0.3887,female,9
                 Uruguay,0.0118,-0.0112,3e-4,male,10
                 Uruguay,-0.3577,-0.4076,-0.3826,female,10")

(p <- uruguay |>
  mutate(pause = if_else(t_es == 0, 2, 1)) |> # set pause for 10 at zero
  uncount(pause) |> # repeat pause rows
  mutate(reveal = row_number(), .by = c(name, gender)) |> # set the reveal sequence
  ggplot(aes(t_es, estimate, group = gender, color = gender)) +
  geom_vline(xintercept = -.5, linetype = "solid", color = "black", linewidth = .2) +
  geom_hline(yintercept = 0, linetype = "solid", color = "black") +
  geom_line(linewidth = 1.5) +
  scale_color_manual(values = c("#D90D0D", "#6A6654")) +
  theme_minimal(base_family = "Verdana",
                base_size = 16) +
  labs(x = "",
       y = "",
       color = "",
       title = "Cambio en el empleo después de tener el primer hijo/a, %",
       subtitle = "Respecto a dos años antes del nacimiento",
       caption = "Fuente: elaboración propia en base a The Child Penalty Atlas, por H. Kleven et al., 2023.\nPaula Pereda Suárez | @paubgood") +
  # 替换所有固定文本为annotate，避免数据绑定导致的错位
  annotate("text", x = -4, y = 0.0100, label = "Varones", fontface = "bold", color = "#6A6654", vjust = -.5, family = "Verdana") +
  annotate("text", x = -4, y = -.03, label = "Mujeres", fontface = "bold", color = "#D90D0D", vjust = -.5, family = "Verdana") +
  annotate("text", x = -.5, y = -.31, label = "-31%", fontface = "bold", color = "#D90D0D", vjust = -.5, family = "Verdana") +
  annotate("text", x = 10, y = -.38, label = "-39%", fontface = "bold", color = "#D90D0D", vjust = -.5, family = "Verdana") +
  annotate("text", x = -4, y = -.39,  label = "5 años antes", fontface = "bold", color = "black",   vjust = -.5) +
  annotate("text", x = 3.5, y = -.39,  label = "10 años después", fontface = "bold", color = "black",   vjust = -.5) +
  theme(text = element_text(family = "Verdana"),
        plot.title = element_text(face = "bold"),
        legend.position = "none",
        panel.grid.major = element_blank(),
        panel.grid.minor = element_blank(),
        axis.text = element_blank(),
        panel.background = element_blank(),
        plot.background  = element_rect(fill  = "#F5F4EE",
                                        color = "#F5F4EE")))

# Save static plot
ggsave("uruguay.png", dpi = dpi_setting, width = plot_width, height = plot_height)

# Create and save the animated plot
anim <- p +
  transition_reveal(along = t_es) +
  enter_fade() + 
  exit_fade() +
  ease_aes('linear') +
  shadow_mark()

# 基于静态图比例自动计算GIF尺寸，保证宽高比严格一致
gif_width <- 1200
gif_height <- round(gif_width * (plot_height / plot_width))

anim_save(
  "motherhood_penalties_uy.gif", 
  animation = anim, 
  end_pause = 10, 
  nframes = 200,  # 增加帧数提升动画流畅度
  fps = 20,       # 合理帧率平衡流畅度与文件大小
  width = gif_width, 
  height = gif_height,
  dpi = dpi_setting,      # 和静态图一致的dpi，保证画质统一
  renderer = magick_renderer(quality = 100)  # 最高质量渲染
)

额外说明 #

• 用annotate()添加固定文本时，所有参数（x/y/label等）都不要放在aes()里，确保文本不会随动画数据变化而移动。
• 若追求更高画质，可继续提升gif_width，但会同步增加文件体积，可根据需求调整。
• 若仍有模糊问题，可尝试降低fps并进一步增加nframes，让每一帧的渲染更精细。

内容的提问来源于stack exchange，提问作者Paula