peer review Vinita

content/post/2021-04-16-functional-programming-with-strava-data/index.Rmd

@@ -57,13 +57,79 @@ There are many variations of passages of Lorem Ipsum available, but the majority
 #reading in data for sandbox
 strav_data <- read_csv(here("content/post/2021-04-16-functional-programming-with-strava-data/data", "strav_csv.csv"))
 
+<<<<<<< Updated upstream
 #starting with some initial tidying
 dataready <- strav_data %>% 
   select(where(~length(unique(.)) > 1)) %>%
   clean_names() %>% 
   mutate_if(is.numeric, ~round(., 2))
 
 
+=======
+#------Vinita------#
+
+#Recommend using rio for importing data
+library(rio)
+library(janitor)
+strav_data = import(here("content/post/2021-04-16-functional-programming-with-strava-data/data", "strav_csv.csv"),setclass = "tibble") %>% 
+                     characterize() %>% 
+                     clean_names()
+#------Vinita------#
+
+
+#initial function for tidying Stravadata
+tidyfunc <- function(strav_data) {
+strav_data %>% 
+  select(where(~length(unique(.)) > 1)) %>%
+  clean_names() %>% 
+  mutate_if(is.numeric, ~round(., 2)) %>% 
+  filter(type != "Swim") %>% 
+  mutate(type = as.factor(type)) %>% 
+  separate(start_date_local, c("date","time"), sep = " ") %>% 
+  select(-start_date)
+}
+
+#Vinita - very cool use of a funtion here. 
+
+#Function taken directly from Lab3 to determine percentage of each activity in dataset
+# Using proportions could be useful to select only the most common activities (e.g., >5%)
+proportions_activity <- function(activities) {
+  map_dbl(split(activities, activities), length) / length(activities)
+}
+
+
+
+#------Vinita------#
+#This is very cool stuff. As a reader, it will be helpful if you add comments as to what's happening at every step
+
+#I am going to try parsing out the following function and please ignore me if I am wrong
+
+tidyfunc <- function(strav_data) {
+strav_data %>% 
+    #select only those columns which do not have columns with just one unique value  - add reason
+  select(where(~length(unique(.)) > 1)) %>%
+    #make those col_names easier to read by adding underscores or making them lower case
+  clean_names() %>% 
+    #round every numeric column  - add reason
+  mutate_if(is.numeric, ~round(., 2)) %>% 
+    #keep all rows except type ="Swim" - add reason
+  filter(type != "Swim") %>% 
+    #convert to factor
+  mutate(type = as.factor(type)) %>% 
+    #parse the start_date_local column into date and time cols; check out lubridate::`parse_date_time()` 
+  separate(start_date_local, c("date","time"), sep = " ") %>% 
+    #select all cols except start_date
+  select(-start_date)
+}
+
+
+#------Vinita------#
+
+
+tidy1 <- tidyfunc(strav_data)
+a <- proportions_activity(tidy1$type)
+
+>>>>>>> Stashed changes
 
 
 # uniquecols <- function(x) length(unique([,y]) > 1)