business-science / correlationfunnel Goto Github PK

I have found this extremely useful as part of initial look at data.

Would it be possible to reduce bin lengths generated, especially where they become recursive, it would make charts much less cluttered.

Regards

For columns that are read as an integer, binarize() is throwing an error.

"Error: binarize(): [Unnacceptable Columns Detected] The following columns contain non-numeric or non-categorical data types"

Let's say I have a continuous response and I'm interested in correlations all along its range, not just the top or bottom quartile.

Instead of running a separate correlation_funnel for each bin, is there any reason I cannot add the continuous column back in between running binarize() and correlate()?

Reprex:

library(dplyr); library(correlationfunnel);

foo <- select(survival::veteran,-'time') %>% binarize() %>% 
    cbind(time=survival::veteran$time) %>% correlate(target=time);

foo$bin[1] <- 'time';

plot_correlation_funnel(foo);

The above runs with one warning and no errors, producing a plot where presumably all correlations are relative to an outcome variable that is not binned just like I want.

My question: Is it valid to use correlationfunel this way?

Thanks.

Great and
very clear stepXstep package tutorial, Matt!.

A time-saving suggestion (if I may):
in Step:
"Examining the Results" (after Step 3),
where you have:

marketing_campaign_correlated_tbl %>%
filter(feature %in% c("DURATION", "POUTCOME", "PDAYS",
"PREVIOUS", "CONTACT", "HOUSING")) %>%
plot_correlation_funnel(interactive = FALSE, limits = c(-0.4, 0.4))

Why not "automatically" generate the results
for these top 6 dependent variables?.
Easy and useful shortcut! :-)

The new, suggested Function:
explain()
where the default is: show only the 6 TOP vars,

The user can specify any other # of top vars to show,
ie:
explain(3) to show the 3 top vars, or
explain(+3) to show the 3 top positively-correlated vars, or
explain(-3) to show the 3 top negatively-correlated vars

Love to see this explain() Fx in correlationFunnel! :-)
THANKS, MATT! great job!.

Sfd99
San Francisco

Hello - When I attempt to use the "Plot a Correlation Funnel" example using the data("marketing_campaign_tbl") dataset. I get the following error:

"Error: plot_correlation_funnel(): [Unnacceptable Data] Acceptable data is generated from the output of correlate()."

I would love to get this code working as it looks to be an invaluable resource.

Here's the reprex:

library(dplyr)
library(correlationfunnel)
marketing_campaign_tbl %>%
select(-ID) %>%
binarize() %>%
correlate(TERM_DEPOSIT__yes) %>%
plot_correlation_funnel()

Session Info:
R version 4.1.2 (2021-11-01)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 19044)

Matrix products: default

locale:
[1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 LC_MONETARY=English_United States.1252
[4] LC_NUMERIC=C LC_TIME=English_United States.1252

attached base packages:
[1] tools grid stats graphics grDevices datasets utils methods base

other attached packages:
[1] reactable_0.2.3 kableExtra_1.3.4.9000 rlang_1.0.6.9000 log4r_0.4.2 zipcodeR_0.3.3
[6] yaml_2.3.5 xtable_1.8-4 xray_0.2.900 xfun_0.31 visdat_0.5.3
[11] validate_1.1.1 UpSetR_1.4.0 tryCatchLog_1.3.1 treemap_2.4-3 timetk_2.8.0
[16] forcats_0.5.2 tidyr_1.2.1 tidyverse_1.3.1 tidytable_0.9.0 taskscheduleR_1.6
[21] summarytools_1.0.0 styler_1.7.0 stringr_1.4.1 splot_0.5.2 snakecase_0.11.0
[26] SmartEDA_0.3.8 SixSigma_0.10.3 skimr_2.1.4 shinyWidgets_0.6.4 shinydashboard_0.7.2
[31] shinyAce_0.4.1 shiny_1.7.1 RVerbalExpressions_0.1.0 rstudio.prefs_0.1.8 rstudioapi_0.14
[36] rmarkdown_2.16 rio_0.5.29 rgdal_1.5-32 sp_1.5-0 reshape2_1.4.4
[41] reprex_2.0.1 ReDaMoR_0.6.3 visNetwork_2.1.0 readxl_1.3.1 readr_2.1.3
[46] raincloudplots_0.2.0 quantmod_0.4.18 TTR_0.24.3 xts_0.12.1 zoo_1.8-9
[51] purrr_0.3.5 processR_0.2.6 ppsr_0.0.2 plotly_4.10.0.9001 plotluck_1.1.1
[56] pivottabler_1.5.3 pdftools_3.1.1 patchwork_1.1.1 pasteAsComment_0.2.0 pak_0.3.1
[61] packagefinder_0.3.2 pacman_0.5.1 orca_1.1-1 openxlsx_4.2.5 naniar_0.6.1.9000
[66] modelsummary_1.0.2.9000 Microsoft365R_2.3.4 magrittr_2.0.3 lubridate_1.8.0 logger_0.2.2
[71] lobstr_1.1.1 lintr_2.0.1 leaflet_2.1.1.9000 knitr_1.40 janitor_2.1.0
[76] inspectdf_0.0.11 inexact_0.0.3 IEDA_0.1.0 htmltools_0.5.3 highcharter_0.9.4.9000
[81] here_1.0.1 gt_0.5.0 grkstyle_0.0.3 googlesheets4_1.0.1.9000 googledrive_2.0.0.9000
[86] ggthemr_1.1.0 ggstatsplot_0.9.1 GGally_2.1.2.9000 ggplot2_3.3.6 ggdogs_1.0
[91] ggblanket_1.0.0 gluedown_1.0.4 fuzzyjoin_0.1.6 futile.logger_1.4.3 fortunes_1.5-4
[96] formattable_0.2.1 formatR_1.12 flow_0.1.0 flextable_0.7.0 flexdashboard_0.5.2
[101] ezknitr_0.6 explore_0.8.0 report_0.5.5 see_0.7.2 correlation_0.8.2
[106] modelbased_0.8.5 effectsize_0.7.0.5 parameters_0.18.2.11 performance_0.9.2 bayestestR_0.13.0
[111] datawizard_0.6.1 insight_0.18.4.6 easystats_0.5.2 dygraphs_1.1.1.7 DT_0.23
[116] dtplyr_1.2.1 dplyr_1.0.9 plyr_1.8.7 downlit_0.4.2 dm_0.2.8.9002
[121] dlookr_0.5.6 diffobj_0.3.5 DiagrammeR_1.0.9 devtools_2.4.3 usethis_2.1.5.9000
[126] deepdep_0.4.1 datapasta_3.1.1 dataReporter_1.0.0 datamodelr_0.2.2.9002 dataMeta_0.1.1
[131] DataExplorer_0.8.2 DataEditR_0.1.5 data.validator_0.1.6 data.tree_1.0.0 DALEX_2.4.2
[136] data.table_1.14.2 daff_0.3.5 d3heatmap_0.9.0 correlationfunnel_0.2.0 compareDF_2.3.3
[141] compare_0.2-6 clock_0.6.0 circlize_0.4.15 checkpoint_1.0.2 cheatsheet_0.1.0
[146] blogsnip_0.0.0.9004 bookdownplus_1.5.8 beepr_1.3 autoEDA_1.0 assertr_2.9
[151] ARTofR_0.4.1 arsenal_3.6.3 archivist_2.3.6 ProjectTemplate_0.10.2 tibble_3.1.7
[156] digest_0.6.29

loaded via a namespace (and not attached):
[1] sjlabelled_1.1.8 mycor_0.1.1 remotes_2.4.2 shinyjs_2.1.0 lattice_0.20-45 paletteer_1.4.0
[7] vctrs_0.4.2.9000 stats4_4.1.2 utf8_1.2.2 blob_1.2.3 R.oo_1.24.0 reactR_0.4.4
[13] withr_2.5.0 foreign_0.8-82 gdtools_0.2.4 uuid_1.1-0 matrixStats_0.61.0 audio_0.1-10
[19] lifecycle_1.0.3.9000 emmeans_1.7.3 cellranger_1.1.0 munsell_0.5.0 ragg_1.2.2 fontawesome_0.2.2
[25] AzureGraph_1.3.2 devEMF_4.0-2 codetools_0.2-18 gghalves_0.1.1 furrr_0.2.3 ppcor_1.1
[31] magick_2.7.3 parallelly_1.32.1 fs_1.5.2 stringi_1.7.6 rlist_0.4.6.2 pbivnorm_0.6.0
[37] pkgconfig_2.0.3 prettyunits_1.1.1 cyclocomp_1.1.0 rvg_0.2.5 estimability_1.3 httr_1.4.4
[43] ggiraphExtra_0.3.0 igraph_1.2.11 progress_1.2.2 hrbrthemes_0.8.0 qpdf_1.1 terra_1.6-17
[49] diagram_1.6.5 haven_2.5.0 mc2d_0.1-21 V8_4.0.0 rsample_0.1.1 miniUI_0.1.1.1
[55] viridisLite_0.4.1 prodlim_2019.11.13 pillar_1.8.1 pkgdown_2.0.3 jquerylib_0.1.4 later_1.3.0
[61] glue_1.6.2 DBI_1.1.2 foreach_1.5.2 ISLR_1.4 robustbase_0.95-0 gtable_0.3.1
[67] raster_3.6-3 tigris_1.6 GlobalOptions_0.1.2 fastmap_1.1.0 extrafont_0.18 sampling_2.9
[73] crosstalk_1.2.0 broom_1.0.1 checkmate_2.1.0 promises_1.2.0.1 webshot_0.5.4 tmvnsim_1.0-2
[79] textshaping_0.3.6 rapportools_1.1 brio_1.1.3 mnormt_2.0.2 hms_1.1.2 askpass_1.1
[85] png_0.1-7 lazyeval_0.2.2 Formula_1.2-4 crayon_1.5.2 extrafontdb_1.0 gridBase_0.4-7
[91] predict3d_0.1.3.3 svglite_2.1.0 flock_0.7 tidyselect_1.1.2 pander_0.6.5 splines_4.1.2
[97] editData_0.1.8 rintrojs_0.3.0 survival_3.2-13 bannerCommenter_1.0.0 rappdirs_0.3.3 WRS2_1.1-3
[103] bit64_4.0.5 lambda.r_1.2.4 modelr_0.1.8 networkD3_0.4 sjmisc_2.8.9 pagedown_0.17
[109] ggsignif_0.6.3 R.methodsS3_1.8.1 rex_1.2.1 markdown_1.1 ggiraph_0.8.2 renv_0.15.4
[115] cachem_1.0.6 ipred_0.9-13 statsExpressions_1.3.0 abind_1.4-5 systemfonts_1.0.4 mime_0.12
[121] ztable_0.2.3 ggrepel_0.9.1 rstatix_0.7.0 processx_3.7.0 xaringan_0.24 interactions_1.1.5
[127] cli_3.4.1 rgl_0.108.3 proxy_0.4-26 future.apply_1.9.1 Matrix_1.3-4 libcoin_1.0-9
[133] shinyBS_0.61 assertthat_0.2.1 officer_0.4.2 repr_1.1.4 lpSolve_5.6.15 mgcv_1.8-38
[139] ggpubr_0.4.0 R.utils_2.12.0 rhandsontable_0.3.8 moonBook_0.3.1 zip_2.2.0 prediction_0.3.14
[145] colourpicker_1.1.1.9001 tzdb_0.3.0 maptools_1.1-2 ps_1.7.1 fansi_1.0.3 KernSmooth_2.23-20
[151] clipr_0.8.0 backports_1.4.1 sysfonts_0.8.5 farver_2.1.1 bit_4.0.4 hardhat_1.2.0
[157] sass_0.4.2.9000 futile.options_1.0.1 partykit_1.2-15 iterators_1.0.14 tables_0.9.6 nlme_3.1-155
[163] lavaan_0.6-11 shape_1.4.6 bslib_0.4.0 inum_1.0-4 sf_1.0-5 rematch2_2.1.2
[169] listenv_0.8.0 gargle_1.2.1.9000 generics_0.1.3 colorspace_2.0-3 base64enc_0.1-3 pkgbuild_1.3.1
[175] e1071_1.7-9 jtools_2.1.4 dbplyr_2.1.1 pryr_0.1.5 RColorBrewer_1.1-3 R.cache_0.15.0
[181] timeDate_4021.106 evaluate_0.16 memoise_2.0.1 coda_0.19-4 semTools_0.5-5 httpuv_1.6.5
[187] class_7.3-20 Rttf2pt1_1.3.10 Rcpp_1.0.8.3 openssl_2.0.3 classInt_0.4-3 pkgload_1.2.4
[193] jsonlite_1.8.2 tidycensus_1.2.1 showtextdb_3.0 bookdown_0.26 rprojroot_2.0.3 bitops_1.0-7
[199] RSQLite_2.2.14 globals_0.16.1 compiler_4.1.2 nnet_7.3-17 settings_0.2.7 tcltk_4.1.2
[205] carData_3.0-5 testthat_3.1.4 rrtable_0.2.1 sessioninfo_1.2.2 lava_1.6.10 ggfittext_0.9.1
[211] rvest_1.0.3 recipes_1.0.1 future_1.28.0 mvtnorm_1.1-3 htmlwidgets_1.5.4 psych_2.2.3
[217] labeling_0.4.2 callr_3.7.2 curl_4.3.3 parallel_4.1.2 highr_0.9 DEoptimR_1.0-11
[223] scales_1.2.1 showtext_0.9-4 desc_1.4.1 gridExtra_2.3 AzureAuth_1.3.3 RCurl_1.98-1.6
[229] car_3.0-12 zeallot_0.1.0 MASS_7.3-55 ellipsis_0.3.2 xml2_1.3.3 gower_1.0.0
[235] reshape_0.8.9 rpart_4.1.16 R6_2.5.1 units_0.7-2

Thank you for your time and consideration.

library(correlationfunnel )

This error occurs while loading the package any arguments need to be passed with loading the packages

Error: package or namespace load failed for ‘correlationfunnel’:
.onAttach failed in attachNamespace() for 'correlationfunnel', details:
call: if (theme$dark) {
error: missing value where TRUE/FALSE needed

Would be nice with something similar as you have for the standard cor-function where users can specify how they would like to deal with NA’s. I do agree with the comment "Missing values and cleaning data are critical to getting great correlations" but a function like this is very convinient when having a few NAs in some columns.