Plot months in x-axis?

[ignacio82]

trafficstars

The ticks in this plot go from 0 to 13:

countspermonth <-
  structure(
    list(
      month = c(4, 5, 6, 7, 8, 9, 10, 11, 12, 1, 2,
                3), count = c(
                  1203409L, 1104115L, 933362L, 940897L, 1149347L,
                  1221000L, 1319375L, 1087002L, 1126721L, 1276824L, 1152835L, 1687903L
                )
    ), row.names = c(NA,-12L), class = "data.frame", .Names = c("month",
                                                                "count")
  )

library(taucharts)
tauchart(countspermonth) %>% 
  tau_line(x = "month", y = "count") %>% 
  tau_guide_x(label="Month") %>% 
  tau_guide_y(label="Counts", label_padding=50)

Is there a way to tell tauchart to use months instead of numbers? Or at least respect the 4,5,..2 ticks?

Thanks!

[hrbrmstr]

perhaps by making them actual dates?

On Wed, Aug 26, 2015 at 9:54 AM, ignacio82 [email protected] wrote:

countspermonth <- structure( list( month = c(4, 5, 6, 7, 8, 9, 10, 11, 12, 1, 2, 3), count = c( 1203409L, 1104115L, 933362L, 940897L, 1149347L, 1221000L, 1319375L, 1087002L, 1126721L, 1276824L, 1152835L, 1687903L ) ), row.names = c(NA,-12L), class = "data.frame", .Names = c("month", "count") ) library(taucharts) tauchart(countspermonth) %>% tau_line(x = "month", y = "count") %>% tau_guide_x(label="Month") %>% tau_guide_y(label="Counts", label_padding=50)

[ignacio82]

Thanks! This this the trick:

countspermonth <-
  structure(
    list(
      month = c(4, 5, 6, 7, 8, 9, 10, 11, 12, 1, 2,
                3), count = c(
                  1203409L, 1104115L, 933362L, 940897L, 1149347L,
                  1221000L, 1319375L, 1087002L, 1126721L, 1276824L, 1152835L, 1687903L
                )
    ), row.names = c(NA,-12L), class = "data.frame", .Names = c("month",
                                                                "count")
  )


countspermonth <- countspermonth %>%  mutate(month = month.abb[month])

library(taucharts)
tauchart(countspermonth) %>% 
  tau_line(x = "month", y = "count") %>% 
  tau_guide_x(label="Month") %>% 
  tau_guide_y(label="Counts", label_padding=50)

[hrbrmstr]

well, this brings up something interesting (poking @timelyportfolio to chime in if he has time, too). Why does:

countspermonth$nam <- month.abb[countspermonth$month]
tauchart(countspermonth) %>% 
  tau_line(x = "nam", y = "count") %>% 
  tau_guide_x(auto_scale=TRUE, label="Month") %>% 
  tau_guide_y(label="Counts", label_padding=50)

work, but:

countspermonth$onam <- factor(countspermonth$month, 
                             levels=countspermonth$month,
                             labels=month.abb[countspermonth$month],
                             ordered=TRUE)

tauchart(countspermonth) %>% 
  tau_line(x = "onam", y = "count") %>% 
  tau_guide_x(auto_scale=TRUE, label="Month") %>% 
  tau_guide_y(label="Counts", label_padding=50)

not (when, IMO it should be the other way around)?

[ignacio82]

I just noticed that my months are out of order :( . Is that fixable?

[hrbrmstr]

Well, that's kind of the issue I just posted & reopened ;-)

On Wed, Aug 26, 2015 at 11:01 AM, ignacio82 [email protected] wrote:

I just noticed that my months are out of order :( . Is that fixable?

— Reply to this email directly or view it on GitHub https://github.com/hrbrmstr/taucharts/issues/49#issuecomment-135051327.

[hrbrmstr]

this works, but i can't figure out how to reduce the number of x axis ticks.

countspermonth$mon <- as.Date(sprintf("%d-%02d-01", c(rep(2013, 9),
                                                      rep(2014, 3)), 
                                      countspermonth$month))

tauchart(countspermonth) %>% 
  tau_line(x = "mon", y = "count") %>% 
  tau_guide_x(auto_scale=TRUE, label="Month", tick_format="%b", tick_period="year") %>% 
  tau_guide_y(label="Counts", label_padding=50)

That's probably a question for the TauCharts team.

[timelyportfolio]

looking at it now

[hrbrmstr]

(totally no rush @timelyportfolio just looping you in on an interesting issue that I think is a TauCharts issue, and not an htmlwidgets one since the data structure up until the handoff to TauCharts.JS is fine)

[timelyportfolio]

Guess it seems like to separate issues to me.

separate from the question posed by @ignacio82 is the auto_scale argument does not flow through in tau_guide_x.

#let's say we wanted 1 to 12 with no 0 or 1
tau <- tauchart(countspermonth) %>% 
  tau_line(x = "month", y = "count") %>% 
  tau_guide_x(label="Month", min = 1, max = 12, auto_scale = FALSE) %>% 
  tau_guide_y(label="Counts", label_padding=50)
tau
#get TRUE
tau$x$guide$x$autoScale

#manually set
tau$x$guide$x$autoScale = TRUE
tau

what we think is the TauCharts problem when we supply an order

[timelyportfolio]

@ignacio82 & @hrbrmstr

For reference, the issue is only with the tau_line probably due to the d3 line constructor. tau_bar and tau_point work as expected.

countspermonth %>%
    mutate( monthorderedfactor = factor(month,ordered =T) ) %>%
    tauchart() %>% 
    tau_bar(x = "monthorderedfactor", y = "count") %>% 
    tau_guide_x(label="Month") %>% 
    tau_guide_y(label="Counts", label_padding=50) %>%
    tau_tooltip

[ignacio82]

@timelyportfolio @hrbrmstr if the dates are yearmon I get this:

countspermonth <-
  structure(
    list(
      yearmon = structure(
        c(
          2014.25, 2014.33333333333,
          2014.41666666667, 2014.5, 2014.58333333333, 2014.66666666667,
          2014.75, 2014.83333333333, 2014.91666666667, 2015, 2015.08333333333,
          2015.16666666667
        ), class = "yearmon"
      ), count = c(
        2492L, 2443L,
        2070L, 2999L, 2468L, 2285L, 3003L, 2467L, 3271L, 2979L, 2581L,
        3601L
      ), requierecrt = c(
        5.17656500802568, 4.25706099058535, 5.16908212560386,
        5.90196732244082, 4.13290113452188, 6.17067833698031, 3.99600399600399,
        7.13417105796514, 4.46346682971568, 5.27022490768715, 6.12165827198761,
        6.2204943071369
      )
    ), .Names = c("yearmon", "count", "crt"), class = "data.frame", row.names = c(NA,-12L)
  )

tauchart(countspermonth) %>% 
  tau_line(x = "yearmon", y = "count") %>% 
  tau_guide_x(label="Month") %>% 
  tau_guide_y(label="Counts", label_padding=50) %>%
  tau_tooltip() %>%
  make_black_tooltip()

Is changing the x axis to just show the year and the month without the day an easy fix?