我有一个数据集df，其中包含102个变量：16个整数，80个因数，8个logi。没有NA值。
我以前使用DataExplorer都没有问题，但是当我在此数据集上运行它时...
library(DataExplorer)
create_report(df)
...进展顺利，输出进度...
# label: correlation_analysis
# |................................................ | 74%
# ordinary text without R code
...直到出现此错误时到达PCA部分：
# |.................................................. | 76%
# label: principle_component_analysis
# Quitting from lines 208-221 (report.rmd)
#
# Error in data.table(pc = paste0("PC", seq_along(pca$sdev)), var = var_exp, :
# Item 2 has no length. Provide at least one item (such as NA, NA_integer_ etc) to be repeated to match the 1 row in the longest column. Or, all columns can be 0 length, for insert()ing rows into.
我搜索了此错误，但只找到解释PCA的页面，而不是此错误的页面。有什么建议吗？
回溯是
26. stop("Item ", i, " has no length. Provide at least one item (such as NA, NA_integer_ etc) to be repeated to match the ",
nr, " row", if (nr > 1L) "s", " in the longest column. Or, all columns can be 0 length, for insert()ing rows into.")
25. data.table(pc = paste0("PC", seq_along(pca$sdev)), var = var_exp,
pct = var_exp/sum(var_exp), cum_pct = cumsum(var_exp)/sum(var_exp))
24. plot_prcomp(data = structure(list(EnrollmentID = c(4603L, 8457L,
3290L, 3323L, 6186L, 6501L, 3084L, 8662L, 7676L, 3229L, 6005L,
3387L, 8204L, 9018L, 4517L, 3320L, 8840L, 7729L, 8835L, 5148L,
7560L, 1239L, 5874L, 4963L, 3755L, 3397L, 9877L, 8609L, 6584L, ...
23. do.call(fun_name, c(list(data = data), report_config[[fun_name]])) at <text>#9
22. do_call("plot_prcomp", na_omit = TRUE) at <text>#8
21. eval(expr, envir, enclos)
20. eval(expr, envir, enclos)
19. withVisible(eval(expr, envir, enclos))
18. withCallingHandlers(withVisible(eval(expr, envir, enclos)), warning = wHandler,
error = eHandler, message = mHandler)
17. handle(ev <- withCallingHandlers(withVisible(eval(expr, envir,
enclos)), warning = wHandler, error = eHandler, message = mHandler))
16. timing_fn(handle(ev <- withCallingHandlers(withVisible(eval(expr,
envir, enclos)), warning = wHandler, error = eHandler, message = mHandler)))
15. valuate_call(expr, parsed$src[[i]], envir = envir, enclos = enclos,
debug = debug, last = i == length(out), use_try = stop_on_error !=
2L, keep_warning = keep_warning, keep_message = keep_message,
output_handler = output_handler, include_timing = include_timing)
14. evaluate::evaluate(...)
13. evaluate(code, envir = env, new_device = FALSE, keep_warning = !isFALSE(options$warning),
keep_message = !isFALSE(options$message), stop_on_error = if (options$error &&
options$include) 0L else 2L, output_handler = knit_handlers(options$render,
options))
12. in_dir(input_dir(), evaluate(code, envir = env, new_device = FALSE,
keep_warning = !isFALSE(options$warning), keep_message = !isFALSE(options$message),
stop_on_error = if (options$error && options$include) 0L else 2L,
output_handler = knit_handlers(options$render, options)))
11. block_exec(params)
10. call_block(x)
9. process_group.block(group)
8. process_group(group)
7. withCallingHandlers(if (tangle) process_tangle(group) else process_group(group),
error = function(e) {
setwd(wd)
cat(res, sep = "\n", file = output %n% "") ...
6. process_file(text, output)
5. knitr::knit(knit_input, knit_output, envir = envir, quiet = quiet,
encoding = encoding)
4. render(input = report_dir, output_file = output_file, output_dir = output_dir,
intermediates_dir = output_dir, params = list(data = data,
report_config = config, response = y), ...)
3. withCallingHandlers(expr, warning = function(w) invokeRestart("muffleWarning"))
2. suppressWarnings(render(input = report_dir, output_file = output_file,
output_dir = output_dir, intermediates_dir = output_dir,
params = list(data = data, report_config = config, response = y),
...))
1. create_report(df)
这是会话信息：
sessionInfo()
R version 3.5.1 (2018-07-02)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows >= 8 x64 (build 9200)
Matrix products: default
locale:
[1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252
[3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C
[5] LC_TIME=English_United States.1252
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] car_3.0-2 knitr_1.20 rmarkdown_1.10 data.table_1.11.8
[5] DataExplorer_0.7.0 mosaic_1.4.0 Matrix_1.2-14 mosaicData_0.17.0
[9] ggformula_0.9.0 ggstance_0.3.1 mdsr_0.1.6 Lahman_6.0-0
[13] ISLR_1.2 forcats_0.3.0 stringr_1.3.1 dplyr_0.7.8
[17] purrr_0.2.5 readr_1.1.1 tidyr_0.8.2 tibble_1.4.2
[21] ggplot2_3.1.0 tidyverse_1.2.1 lattice_0.20-35 carData_3.0-2
loaded via a namespace (and not attached):
[1] ggdendro_0.1-20 httr_1.3.1 RMySQL_0.10.15 jsonlite_1.5 splines_3.5.1
[6] modelr_0.1.2 assertthat_0.2.0 highr_0.7 cellranger_1.1.0 yaml_2.2.0
[11] ggrepel_0.8.0 pillar_1.3.0 backports_1.1.2 glue_1.3.0 downloader_0.4
[16] digest_0.6.18 rvest_0.3.2 colorspace_1.3-2 htmltools_0.3.6 plyr_1.8.4
[21] pkgconfig_2.0.2 broom_0.5.0 haven_1.1.2 scales_1.0.0 openxlsx_4.1.0
[26] rio_0.5.10 withr_2.1.2 lazyeval_0.2.1 cli_1.0.1 magrittr_1.5
[31] crayon_1.3.4 readxl_1.1.0 evaluate_0.12 nlme_3.1-137 MASS_7.3-50
[36] xml2_1.2.0 foreign_0.8-71 tools_3.5.1 hms_0.4.2 munsell_0.5.0
[41] babynames_0.3.0 zip_1.0.0 bindrcpp_0.2.2 networkD3_0.4 compiler_3.5.1
[46] rlang_0.3.0.1 grid_3.5.1 rstudioapi_0.8 htmlwidgets_1.3 igraph_1.2.2
[51] labeling_0.3 mosaicCore_0.6.0 gtable_0.2.0 abind_1.4-5 DBI_1.0.0
[56] curl_3.2 reshape2_1.4.3 R6_2.3.0 gridExtra_2.3 lubridate_1.7.4
[61] rprojroot_1.3-2 bindr_0.1.1 stringi_1.2.4 parallel_3.5.1 Rcpp_1.0.0
[66] dbplyr_1.2.2 tidyselect_0.2.5
下面是注释中所要求的Introduction（df_dummified）的输出：
A tibble: 1 x 9
rows columns discrete_columns continuous_columns
<int> <int> <int> <int>
9527 489 2 487
all_missing_columns total_missing_values
<int> <int>
0 7826
complete_rows total_observations memory_usage
<int> <int> <dbl>
6889 4658703 18919440
答案 0 :(得分：1)
PCA只能应用于数字数据。仅考虑PCA的数字列，除去数字以外的列。
nums <- unlist(lapply(df, is.numeric))
df_new <- df[, nums]
删除所有具有恒定方差的列。
df_new <- df_new[, apply(df_new, 2, var) != 0]
参考：How to solve prcomp.default(): cannot rescale a constant/zero column to unit variance
现在，运行此命令。这应该为您创建一个不错的html报告。
create_report(df_new)
答案 1 :(得分：1)
您还可以考虑通过从create_report（）配置中删除“ plot_prcomp”来跳过报告的PCA部分。
我遇到了同样的问题，但这仍然为我创建了报告的其余部分：
library(DataExplorer)
config <- list(
"introduce" = list(),
"plot_str" = list(
"type" = "diagonal",
"fontSize" = 35,
"width" = 1000,
"margin" = list("left" = 350, "right" = 250)
),
"plot_missing" = list(),
"plot_histogram" = list(),
"plot_qq" = list(sampled_rows = 1000L),
"plot_bar" = list(),
"plot_correlation" = list("cor_args" = list("use" = "pairwise.complete.obs")),
# "plot_prcomp" = list(),
"plot_boxplot" = list(),
"plot_scatterplot" = list(sampled_rows = 1000L)
)
create_report(df, config = config)