Thanks so much for this project, I found it immensely helpful. I wanted to suggest a m

I ultimately went with a <a href="https://github.com/ASKurz/Statistical_Rethinking_wit

Simplification of figure 12.2.a about statistical_rethinking_with_brms_ggplot2_and_the_tidyverse HOT 5 CLOSED

askurz commented on May 30, 2024

Simplification of figure 12.2.a

from statistical_rethinking_with_brms_ggplot2_and_the_tidyverse.

Comments (5)

ASKurz commented on May 30, 2024 1

That’s a great suggestion! I played around with different solutions to that kind of plot in a recent blog and this is a nice addition to those methods. Thanks for bringing it to my attention. Before I release the next revision of the Rethinking project, I’ll update this plot, for sure.

from statistical_rethinking_with_brms_ggplot2_and_the_tidyverse.

ASKurz commented on May 30, 2024

I ultimately went with a different solution. But the spirit is the same. It'll appear in the upcoming bookdown update. Thanks again for opening the issue.

from statistical_rethinking_with_brms_ggplot2_and_the_tidyverse.

gglee4ai commented on May 30, 2024

There are some zigzags in Fig. 12.2.a. I am not sure what is the reason. But the set.seed and sample_n affected the pattern. Please check the figure.

from statistical_rethinking_with_brms_ggplot2_and_the_tidyverse.

ASKurz commented on May 30, 2024

What seed values gave you noticeably different results?

from statistical_rethinking_with_brms_ggplot2_and_the_tidyverse.

gglee4ai commented on May 30, 2024

I reproduced the graph in my PC.
I guess there is a problem in (sample_n, expand, nesting)

Fig. 12.2.b with zig-zag lines

sample_n: 35

set.seed(12)
post %>%
  sample_n(35, replace = FALSE) %>%
  expand(nesting(iter, b_Intercept, sd_tank__Intercept),
    x = seq(-4, 5, length.out = 100)
  ) %>%
  mutate(
    y = dnorm(x, b_Intercept, sd_tank__Intercept)
  ) %>%
  ggplot(aes(x = x, y = y, group = iter)) +
  geom_line(alpha = .5, color = "orange2") +
  labs(
    title = "Population survival distribution",
    subtitle = "The Gaussians are on the log-odds scale."
  ) +
  scale_y_continuous(NULL, breaks = NULL) +
  coord_cartesian(xlim = c(-3, 4)) +
  theme_fivethirtyeight() +
  theme(
    plot.title = element_text(size = 13),
    plot.subtitle = element_text(size = 10)
  )

https://raw.githubusercontent.com/gglee4ai/public/master/sample35.png

Fig. 12.2.b without zig-zag lines

sample_n: 30

set.seed(12)
post %>%
  sample_n(30, replace = FALSE) %>%
  expand(nesting(iter, b_Intercept, sd_tank__Intercept),
    x = seq(-4, 5, length.out = 100)
  ) %>%
  mutate(
    y = dnorm(x, b_Intercept, sd_tank__Intercept)
  ) %>%
  ggplot(aes(x = x, y = y, group = iter)) +
  geom_line(alpha = .5, color = "orange2") +
  labs(
    title = "Population survival distribution",
    subtitle = "The Gaussians are on the log-odds scale."
  ) +
  scale_y_continuous(NULL, breaks = NULL) +
  coord_cartesian(xlim = c(-3, 4)) +
  theme_fivethirtyeight() +
  theme(
    plot.title = element_text(size = 13),
    plot.subtitle = element_text(size = 10)
  )

https://raw.githubusercontent.com/gglee4ai/public/master/sample30.png

from statistical_rethinking_with_brms_ggplot2_and_the_tidyverse.

Simplification of figure 12.2.a about statistical_rethinking_with_brms_ggplot2_and_the_tidyverse HOT 5 CLOSED

Comments (5)

Fig. 12.2.b with zig-zag lines

Fig. 12.2.b without zig-zag lines

Related Issues (20)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent