kyle.belanger/DHSC-Capstone
1
0
Fork 0
Code Issues 1 Pull requests Projects Releases Packages Wiki Activity Actions

Update 1-data-exploration.R

Browse source
This commit is contained in:
Kyle Belanger 2023-01-11 12:48:02 -05:00
parent 2b2e148eab
commit c6f864cc48
1 changed files with 16 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

22
ML/1-data-exploration.R
View file

@ -12,7 +12,7 @@ box::use(
,dplyr
,readr
,tidyr
,ggplot2
,gp2 = ggplot2[ggplot, aes]
)
@ -62,13 +62,23 @@ ds_high_tsh <- ds_high_tsh_raw %>%
# basic visualization -----------------------------------------------------
test <- dplyr$as_tibble(colSums(is.na(ds_high_tsh)), rownames = NA ) %>%
tibble::rownames_to_column()
g_count <- dplyr$as_tibble(colSums(is.na(ds_high_tsh)), rownames = NA ) %>%
tibble::rownames_to_column() %>%
ggplot(aes(x = rowname, y = value)) +
gp2$geom_col() +
gp2$theme(
axis.text.x = gp2$element_text(angle = 90)
)
g_count
#quick recode of gender, will still do recoding during feature engineering
g1 <- ds_high_tsh %>%
dplyr$select(-subject_id, - charttime) %>%
dplyr$mutate(dplyr$across(gender, ~dplyr$recode(.,M = 1, F = 2))) %>%
tidyr$pivot_longer(cols = dplyr$everything())
tidyr$pivot_longer(cols = dplyr$everything()) %>%
ggplot(aes(x = value)) +
gp2$geom_histogram() +
gp2$facet_wrap(~name, scales = "free")
g1