kyle.belanger/DHSC-Capstone
1
0
Fork 0
Code Issues1 Pull requests Projects Releases Packages Wiki Activity Actions

Update 1-data-exploration.R

Browse source
This commit is contained in:
Kyle Belanger 2023-01-13 14:27:55 -05:00
parent 4b794c104d
commit ff36565211
1 changed files with 3 additions and 3 deletions

6
ML/1-data-exploration.R
View file

@ -62,6 +62,7 @@ ds_high_tsh <- ds_high_tsh_raw %>%
# basic visualization ----------------------------------------------------- # basic visualization -----------------------------------------------------
# count of missing values
g_count <- dplyr$as_tibble(colSums(is.na(ds_high_tsh)), rownames = NA ) %>% g_count <- dplyr$as_tibble(colSums(is.na(ds_high_tsh)), rownames = NA ) %>%
tibble::rownames_to_column() %>% tibble::rownames_to_column() %>%
ggplot(aes(x = rowname, y = value)) + ggplot(aes(x = rowname, y = value)) +
@ -72,6 +73,7 @@ g_count <- dplyr$as_tibble(colSums(is.na(ds_high_tsh)), rownames = NA ) %>%
g_count g_count
#corrleation plot
ds_corr <- cor(ds_high_tsh %>% dplyr$select(-subject_id, - charttime) ds_corr <- cor(ds_high_tsh %>% dplyr$select(-subject_id, - charttime)
%>% dplyr$mutate(dplyr$across(gender, ~dplyr$recode(.,M = 1, F = 2))) %>% dplyr$mutate(dplyr$across(gender, ~dplyr$recode(.,M = 1, F = 2)))
,use = "complete.obs") ,use = "complete.obs")
@ -86,9 +88,7 @@ dev.off()
# quick view of distribution of features
#quick recode of gender, will still do recoding during feature engineering #quick recode of gender, will still do recoding during feature engineering
g1 <- ds_high_tsh %>% g1 <- ds_high_tsh %>%
dplyr$select(-subject_id, - charttime) %>% dplyr$select(-subject_id, - charttime) %>%