Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Added fable/feast support #2045

Open
wants to merge 3 commits into
base: master
Choose a base branch
from

Conversation

moutikabdessabour
Copy link
Contributor

@moutikabdessabour moutikabdessabour commented Oct 13, 2021

Implemented to_basic methods that handle the output of fabletools:autoplot.fbl_ts and Added tests

library(dplyr)
library(forecast)
library(tsibble)

data <- tourism %>%
    filter(Region == "Melbourne") %>%
    `[`(, c("Quarter", "Trips", "Region")) %>%
    distinct(Quarter, .keep_all = TRUE) %>%
    as_tsibble(key = Region) 
p <- data %>%
    model(
        ets = ETS(Trips ~ trend("A")),
    ) %>%
    forecast(h = "5 years") %>%
    autoplot(data)

p

ggplotly(p)

ggplot

ggplotly
.

Comment on lines +648 to +650
to_basic.data.frame <- function(data, prestats_data, layout, params, p, ...) {
prefix_class(data, "GeomPath")
}
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm, this doesn't seem right, do you have an example where this becomes relevant?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this is actually a trick because the autoplot returns only a data.frame and GeomHilo... this results in only drawing the predictions without the line.

Comment on lines +620 to +641
transform_hiloribbon <- function(data) {
data <- data[order(data$x), ]
data$hilo <- NULL

data$x_plotlyDomain <- as.character(data$x_plotlyDomain)

maximum_lev <- max(data$level) + 1

data$alpha <- (maximum_lev * (maximum_lev - data$level) - 1 )/ maximum_lev**3
data$colour <- data$alpha

unused_aes <- ! names(data) %in% c("x", "y", "ymin", "ymax")

row_number <- nrow(data)

data_rev <- data[row_number:1L, ]
structure(rbind(
cbind(x = data$x, y = data$ymin, data[unused_aes]),
cbind(x = data$x[row_number], y = data$ymin[row_number], data[row_number, unused_aes]),
cbind(x = data_rev$x, y = data_rev$ymax, data_rev[unused_aes])
), class = class(data))
}
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Come to think of it, I think most of this as well as to_basic.GeomAlluvium() could be replaced by ribbon_dat()?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It's the same code with minor changes.

@cpsievert
Copy link
Collaborator

@danton267 would you mind filing an issue to request this feature so that it should up in this list of issues? https://github.com/plotly/plotly.R/issues/created_by/danton267

@cpsievert cpsievert mentioned this pull request Nov 2, 2021
26 tasks
@cpsievert
Copy link
Collaborator

@moutikabdessabour what package is the model() function from in your example?

@cpsievert
Copy link
Collaborator

@moutikabdessabour if we're going to add support for {fable}/{feasts}, we should try our best to cover anything that one could create via forecast::forecast() + autoplot(), which would at least include forecast::GeomForecast

@moutikabdessabour
Copy link
Contributor Author

@cpsievert Ok I will work on it.

@moutikabdessabour
Copy link
Contributor Author

@cpsievert The model function is exported from the {fabletools} package

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants