diff --git a/read_wapData.R b/read_wapData.R
index 693efa28..7cd0451b 100644
--- a/read_wapData.R
+++ b/read_wapData.R
@@ -50,7 +50,7 @@ rpi_wap_stats <- readRDS("../COVID_RPI_WiFi_Data/rpi_wifi_semester_day_summary.r
 #buildinginfo <- readRDS("../COVID_RPI_WiFi_Data/buildinginfo.rds")
 
 #user_prediction: Building, weekday, Hour, users, Mean_Usercount, latitude, longitude, buildingType
-user_predictions <- readRDS("../COVID_RPI_WiFi_Data/rpi_user_predictions.rds")
+user_predictions <- readRDS("../COVID_RPI_WiFi_Data/median_last3wks_with_floors.rds")
 
 ################################################################################################################
 ###CLEANING DATA
@@ -81,12 +81,13 @@ hits_per_wap_semester_by_building_max <- hits_per_wap_semester_by_building_max %
 hits_per_wap_semester_by_building_max <- hits_per_wap_semester_by_building_max %>% group_by(Building) %>% summarise_all(funs(max)) %>% ungroup() %>% select(Building, usercount_max)
 colnames(hits_per_wap_semester_by_building_max) <- c('Building', 'capacity')
 
+user_predictions <- user_predictions %>% 
+  dplyr::group_by(Building, Weekday, Hour, lat, lng, BuildingType) %>%
+  dplyr::summarize(totalusers = sum(users)) %>%
+  ungroup()
 
-#user_predictions <- user_predictions %>% mutate(Hour = hour(as.POSIXct(time))) %>% select(Building, Median_Usercount, Mean_Usercount, Hour, latitude, longitude, buildingType, weekday)
-user_predictions <- user_predictions %>% group_by(Building, Hour, latitude, longitude, buildingType, weekday) %>% ungroup() 
-colnames(user_predictions) <- c('Building', 'weekday', 'Hour','users', 'Mean_Usercount' ,'lat','lng', 'BuildingType')
-user_predictions <- user_predictions %>% filter(BuildingType != "housing" & BuildingType != "greek") %>%
-  filter(!(Building %in% remove_list))
+names(user_predictions)[names(user_predictions) == 'Weekday'] <- 'weekday'
+names(user_predictions)[names(user_predictions) == 'totalusers'] <- 'users'
 
 ################################################################################################################
 ## DEFININING LISTS AND DATA FRAMES FOR CONVENIENCE