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 2023-01-12 15:30:08 -05:00
parent edcfcdab55
commit b863b6304d
1 changed files with 7 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
ML/1-data-exploration.R
View file

@ -72,6 +72,12 @@ g_count <- dplyr$as_tibble(colSums(is.na(ds_high_tsh)), rownames = NA ) %>%
g_count g_count
ds_corr <- cor(ds_high_tsh %>% dplyr$select(-subject_id, - charttime)
%>% dplyr$mutate(dplyr$across(gender, ~dplyr$recode(.,M = 1, F = 2)))
,use = "complete.obs")
corrplot::corrplot(ds_corr, method = "number")
#quick recode of gender, will still do recoding during feature engineering #quick recode of gender, will still do recoding during feature engineering
g1 <- ds_high_tsh %>% g1 <- ds_high_tsh %>%
@ -91,7 +97,7 @@ g2 <- ds_high_tsh %>%
tidyr$pivot_longer(cols = !ft4_dia) %>% tidyr$pivot_longer(cols = !ft4_dia) %>%
ggplot(aes(x = factor(ft4_dia), y = value, fill = factor(ft4_dia))) + ggplot(aes(x = factor(ft4_dia), y = value, fill = factor(ft4_dia))) +
gp2$geom_boxplot(outlier.shape = NA, na.rm = TRUE) + gp2$geom_boxplot(outlier.shape = NA, na.rm = TRUE) +
gp2$geom_jitter(size=.7, width=.1, alpha=.5) + gp2$geom_jitter(size=.7, width=.1, alpha=.5, na.rm = TRUE) +
gp2$facet_wrap(~name, scales = "free") gp2$facet_wrap(~name, scales = "free")
g2 g2