vis_value()for visualising all values in a dataset. It rescales values to be between 0 and 1. See #100vis_binary()for visualising datasets with binary values - similar tovis_value(), but just for binary data (0, 1, NA). See #125. Thank you to Trish Gilholm for her suggested use case for this.- Implemented facetting in
vis_dat()andvis_cor(), andvis_miss()see (#78). The next release will implement facetting forvis_value(),vis_binary(),vis_compare(),vis_expect(), andvis_guess(). - Implemented data methods for plots with
data_vis_dat(),data_vis_cor(), anddata_vis_miss()see (#78). vis_dat()vis_miss()andvis_guess()now render missing values in list-columns (@cregouby #138)- Added
abbreviate_vars()function to assist with abbreviating data names (#140) - Percentage missing in columns for
vis_miss()is now rounding to integers - for more accurate representation of missingness summaries please use thenaniarR package. - A new vignette on customising colour palettes in visdat, "Customising colour palettes in visdat".
- no longer use old version of
gather_(#141) - resolve bug where
vis_value()displayed constant values as NA values (#128) - these constant values are now shown as 1. - removed use of the now deprecated "aes_string" from ggplot2
- output of plot in
vis_expectwould reorder columns (#133), fixed in #143 by @muschellij2. vis_miss()displayed missing percentages between 0.1% and 0.5% as 0% due to rounding. Now it dislpays "<1%" by @zeehio at #162.
- No longer uses gdtools for testing (#145)
- Use
cliinternally for error messages. - Speed up some internal functions in visdat
- Update
vis_cor()to use perceptually uniform colours fromscicopackage, usingscico::scico(3, palette = "vik"). - Update
vis_cor()to have fixed legend values from -1 to +1 (#110) using optionsbreaksandlimits. Special thanks to this SO thread for the answer - Uses
glueandglue_collapse()instead ofpasteandpaste0 - adds WORDLIST for spelling thanks to
usethis::use_spell_check()
- Jim Hester fixed recent changes in readr 1.2.0 in PR #103, which changes the default behavior of the
guess_parser, to not guess integer types by default. To opt-into the current behavior you need to passguess_integer = TRUE.
vis_compare()for comparing two dataframes of the same dimensionsvis_expect()for visualising where certain values of expectations occur in the data- Added NA colours to
vis_expect - Added
show_percarg tovis_expectto show the percentage of expectations that are TRUE. #73
- Added NA colours to
vis_corto visualise correlations in a dataframevis_guess()for displaying the likely type for each cell in a dataframe- Added draft
vis_expectto make it easy to look at certain appearances of numbers in your data. - visdat is now under the rOpenSci github repository
- added CITATION for visdat to cite the JOSS article
- updated options for
vis_corto use argumentna_actionnotuse_op. - cleaned up the organisation of the files and internal functions
- Added appropriate legend and x axis for
vis_miss_ly- thanks to Stuart Lee - Updated the
paper.mdfor JOSS - Updated some old links in doco
- Added Sean Hughes and Mara Averick to the DESCRIPTION with
ctb. - Minor changes to the paper for JOSS
-
Fix bug reported in #75 where
vis_dat(diamonds)erroredseq_len(nrow(x))inside internal functionvis_gather_, used to calculate the row numbers. Usingmutate(rows = dplyr::row_number())solved the issue. -
Fix bug reported in #72 where
vis_misserrored when one column was given to it. This was an issue with usinglimitsinsidescale_x_discrete- which is used to order the columns of the data. It is not necessary to order one column of data, so I created an if-else to avoid this step and return the plot early. -
Fix visdat x axis alignment when show_perc_col = FALSE - #82
-
fix visdat x axis alignment - issue 57
-
fix bug where the column percentage missing would print to be NA when it was exactly equal to 0.1% missing. - issue 62
-
vis_cordidn't gather variables for plotting appropriately - now fixed
- lightweight CRAN submission - will only contain functions
vis_datandvis_miss
add_vis_dat_pal()(internal) to add a palette forvis_datandvis_guessvis_guessnow gets a palette argument likevis_dat- Added protoype/placeholder functions for
plotlyvis_*_ly interactive graphs:vis_guess_ly()vis_dat_ly()vis_compare_ly()These simply wrapplotly::ggplotly(vis_*(data)). In the future they will be written inplotlyso that they can be generated much faster
- corrected testing for
vis_*family - added .svg graphics for correct vdiffr testing
- improved hover print method for plotly.
- axes in
vis_family are now flipped by default vis_missnow shows the % missingness in a column, can be disabled by settingshow_perc_colargument to FALSE- removed
flipargument, as this should be the default
- added internal functions to improve extensibility and debugging -
vis_create_,vis_gather_andvis_extract_value_. - suppress unneeded warnings arising from compiling factors
- Added testing for visualisations with
vdiffr. Code coverage is now at 99% - Fixed up suggestions from
goodpractice::gp() - Submitted to rOpenSci onboarding
paper.mdwritten and submitted to JOSS
- Added feature
flip = TRUE, tovis_datandvis_miss. This flips the x axis and the ordering of the rows. This more closely resembles a dataframe. vis_miss_lyis a new function that uses plotly to plot missing data, likevis_miss, but interactive, without the need to callplotly::ggplotlyon it. It's fast, but at the moment it needs a bit of love on the legend front to maintain the style and features (clustering, etc) of currentvis_miss.vis_missnow gains ashow_percargument, which displays the % of missing and complete data. This is switched on by default and addresses issue #19.
vis_compareis a new function that allows you to compare two dataframes of the same dimension. It gives a fairly ugly warning if they are not of the same dimension.vis_datgains a "palette" argument in line with issue 26, drawn from http://colorbrewer2.org/, there are currently three arguments, "default", "qual", and "cb_safe". "default" provides the ggplot defaults, "qual" uses some colour blind unfriendly colours, and "cb_safe" provides some colours friendly for colour blindness.
- All lines are < 80 characters long
- removed all instances of
1:rnow(x)and replaced withseq_along(nrow(x)). - Updated documentation, improved legend and colours for
vis_miss_ly. - removed export for
vis_dat_ly, as it currently does not work. - Removed a lot of unnecessary @importFrom tags, included magrittr in this, and added magrittr to Imports
- Changes ALL CAPS Headers in news to Title Case
- Made it clear that
vis_guess()andvis_compareare very beta - updated documentation in README and
vis_dat(),vis_miss(),vis_compare(), andvis_guess() - updated pkgdown docs
- updated DESCRIPTION URL and bug report
- Changed the default colours of
vis_compareto be different to the ggplot2 standards. vis_misslegend labels are created using the internal functionmiss_guide_label.miss_guide_labelwill check if data is 100% missing or 100% present and display this in the figure. Additionally, if there is less than 0.1% missing data, "<0.1% missingness" will also be displayed. This sort of gets around issue #18 for the moment.- tests have been added for the
miss_guide_labellegend labels function. - Changed legend label for
vis_miss,vis_dat, andvis_guess. - updated README
- Added vignette folder (but not vignettes added yet)
- Added appveyor-CI and travis-CI, addressing issues #22 and #23
- Update
vis_dat()to usepurrr::dmap(fingerprint)instead ofmutate_each_(). This solves issue #3 wherevis_datcouldn't take variables with spaces in their name.
=========================
- Interactivity with
plotly::ggplotly! Funcionsvis_guess(),vis_dat(), andvis_misswere updated so that you can make them all interactive using the latest dev version ofplotlyfrom Carson Sievert.
=========================
- Introducing
vis_guess(), a function that uses the unexported functioncollectorGuessfromreadr.
=========================
vis_miss()andvis_datactually run