.Rhistory

group_by(contract_type) %>%
summarize(table(contract_type))
df_new %>%
group_by(event_type) %>%
summarize(table(event_type))
df_new %>%
group_by(contract_type) %>%
summarize(table(contract_type))
df_new %>%
group_by(event_type) %>%
summarize(table(event_type))
table(df_new$contract_type)
table(df_new$event_type)
?table
table(df[,c("contract_type", "event_type")])
df[,c("contract_type", "event_type")]
df[,c("contract_type")]
table(df_new[,c("contract_type", "event_type")])
table(df_new[,c("contract_type", "event_type")])
table(df_new[df_new$event_type == "rebill", $rebill_sequence])
table(df_new[df_new$event_type == "rebill", "rebill_sequence"])
head(
df_new[df_new$event_type == "rebill", "rebill_sequence"]
)
summary(
df_new[df_new$event_type == "rebill", "rebill_sequence"]
)
table(df_new[,c("contract_type", "event_type")])
df_ending = runQueryBI("
SELECT
*
FROM
bi.dim_contract
WHERE
contract_end_date > NOW() - INTERVAL 1 MONTH
", db="bi")
table(df_ending[df_new$event_type == "rebill", "rebill_sequence"])
boxplot(df_ending[df_new$event_type == "rebill", "rebill_sequence"])
library(nippon)
install.packages("nippon")
install.packages("Nippon")
library(nippon)
library(Nippon)
gpclibPermit()
if (require(RColorBrewer)) {
cols <- rev(brewer.pal(12,"Set2"))
}else{
cols <- gray.colors(12)
}
JapanPrefecturesMap(col = cols, axes = TRUE)
source('~/.active-rstudio-document', echo=TRUE)
remove.packages("choroplethr")
install.packages("choroplethr")
library(choroplethr)
library(ggplot2)
library(rcolorbrewer)
choroplethr(df_pop_zip, "zip") + scale_fill_brewer(palette=2)
choroplethr(df_pop_zip, "zip") + scale_fill_brewer(palette=2)
choroplethr(df_pop_zip, "zip") + scale_fill_brewer(palette=3)
choroplethr(df_pop_county, "county") + scale_fill_brewer(palette=2)
choroplethr(df_pop_state, "state") + scale_fill_brewer(palette=2)
df_pop_state[2,2] = df_pop_state[2,2]*100
choroplethr(df_pop_state, "state") + scale_fill_brewer(palette=2)
choroplethr(df_pop_zip, "zip") + scale_color_brewer(palette=2)# scale_fill_brewer(palette=2)
source('~/.active-rstudio-document', echo=TRUE)
source('~/.active-rstudio-document', echo=TRUE)
nrow(df)
library(dplyr)
update.packages()
library(dplyr)
head(df)
t = table(df$t)
boxplot(t)
hist(t)
?summarize
df_2 = summarize(df, t, n=count())
df_2 = group_by(df, t) summarize(df, t, n=count())
df_2 = group_by(df, t)
df_3 = summarize(df_2, n=count())
df_3 = summarize(df_2, count = n())
df_3
?geom_hist
?geom_bar
ggplot(df_3, aes(x=t, y=count)) + geom_hist(stat="identity")
ggplot(df_3, aes(x=t, y=count)) + geom_bar(stat="identity")
df_3
df %>%
group_by(orig_contract_config) %>%
head()
df %>%
group_by(orig_contract_config) %>%
summarize(count=n())
df_big_hist = df %>%
group_by(t) %>%
summarize(df, count=n())
ggplot(df_big_hist, aes(x=t, y=count)) +
geom_bar(stat="identity") + # looks like the 1 case of 13 is being dropped....
ggtitle("Histogram of months in this dataset")
df_big_hist = df %>%
group_by(t) %>%
summarize(df, count=n())
head(df)
df %>%
group_by(t) %>%
summarize(df, count=n())
df_big_hist = df %>%
group_by(t) %>%
summarize(count=n())
ggplot(df_big_hist, aes(x=t, y=count)) +
geom_bar(stat="identity") + # looks like the 1 case of 13 is being dropped....
ggtitle("Histogram of months in this dataset")
df_big_hist = df %>%
group_by(t) %>%
summarize(count=n())
ggplot(df_big_hist, aes(x=t, y=count)) +
geom_bar(stat="identity") + # looks like the 1 case of 13 is being dropped....
ggtitle("Histogram of months in this dataset")
df_big_hist = df %>%
group_by(t) %>%
summarize(count=n())
ggplot(df_big_hist, aes(x=t, y=count)) +
geom_bar(stat="identity") + # looks like the 1 case of 13 is being dropped....
ggtitle("Histogram of months in this dataset")
ggplot(df_big_hist, aes(t)) + geom_bar()
ggplot(df_big_hist, aes(t)) + geom_bar(stat="identity")
df_big_hist
ggplot(df_big_hist, aes(t)) + geom_bar()
ggplot(df_big_hist, aes(t)) + geom_bar(stat="identity")
df %>%
group_by(t) %>%
summarize(count=n()) %>%
ggplot(aes(x=t, y=count)) +
geom_bar(stat="identity") + # looks like the 1 case of 13 is being dropped....
ggtitle("Histogram of months in this dataset")
df %>%
group_by(orig_contract_config) %>%
summarize(count=n()) %>%
ggplot(aes(x=orig_contract_config, y=count))
df %>%
group_by(orig_contract_config) %>%
summarize(count=n())
df %>%
group_by(orig_contract_config) %>%
summarize(count=n()) %>%
ggplot(aes(x=orig_contract_config, y=count)) +
geom_bar(stat="identity") +
ggtitle("Original Contract Configuration Frequency")
?df_president_ts
library(choroplethr)
?df_president_ts
make_histogram = function(df, t)
{
df %>%
group_by(t) %>%
summarize(count=n()) %>%
ggplot(aes(x=t, y=count)) +
geom_bar(stat="identity") +
ggtitle(t)
}
make_histogram(df, "all")
head(df_2)
df=df %>% group_by(orig_contract_config) %>% summarize(count=n())
head(df)
df = runQueryBI("SELECT * FROM analytics.dtsm");
rm(df_2)
rm(df_3)
rm(df_big_hist)
d=df %>% group_by(t) %>% summarize(count=n())
d
d$count = d$count/d$count[1]
d
# for modeling or visualization, remove periods that are in the future
future_filter <- function (df) {
filter(df, period_end < Sys.Date() - 1)
}
# to build a life-table from the data
life_table <- function (df, condition) {
condition <- c(condition, 't')
group_var <- lapply(condition, as.symbol)
regroup(df, group_var) %>%
summarize(cases = n(), cancels = sum(cancel), hazard = cancels / cases, period_surv = 1 - hazard) %>%
mutate(survival = order_by(t, cumprod(period_surv)))
}
head(df)
life_table_df <- life_table(future_filter(df), 'orig_contract_config')
p <- ggplot(life_table_df, aes(x = t, y = survival, color=orig_contract_config)) + geom_line()
p
?filter
head(life_table_df)
life_table_df
source('~/.active-rstudio-document', echo=TRUE)
source('~/.active-rstudio-document', echo=TRUE)
head(df)
bak=d
df$d=as.POSIXlt(df$d)
ggplot(df, aes(d, num_over_36)) + geom_line()
ggplot(df, aes(d, num_over_36)) + geom_line() + ggitle("Number of emails over Weight 36 each day\nlast 2 years")
ggplot(df, aes(d, num_over_36)) + geom_line() + ggtitle("Number of emails over Weight 36 each day\nlast 2 years")
ggplot(df, aes(d, num_over_36)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails")
q = "
SELECT
d, COUNT(*) num_over_15
FROM
daily_email_weights
WHERE
weight > 15
GROUP BY
1
"
df = runQuery(q, "LeadInbox")
head(df)
bak=d
df$d=as.POSIXlt(df$d)
ggplot(df, aes(d, num_over_15)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails")
rm(d)
rm(q)
q_36 = "
SELECT
d, COUNT(*) num_over_36
FROM
daily_email_weights
WHERE
weight > 36
GROUP BY
1
"
df_36 = runQuery(q_36, "LeadInbox")
df$d=as.POSIXlt(df_36$d)
ggplot(df, aes(df_36, num_over_36)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails")
q_36 = "
SELECT
d, COUNT(*) num_over_36
FROM
daily_email_weights
WHERE
weight > 36
GROUP BY
1
"
df_36 = runQuery(q_36, "LeadInbox")
df_36$d=as.POSIXlt(df_36$d)
ggplot(df_36, aes(df_36, num_over_36)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails")
ggplot(df_36, aes(df, num_over_36)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails")
ggplot(df_36, aes(d, num_over_36)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails")
q_15 = "
SELECT
d, COUNT(*) num_over_15
FROM
daily_email_weights
WHERE
weight > 15
GROUP BY
1
"
df_15 = runQuery(q, "LeadInbox")
df_15 = runQuery(q_15, "LeadInbox")
df_15 = runQuery(q_15, "LeadInbox")
df_15$d=as.POSIXlt(df_15$d)
ggplot(df_15, aes(d, num_over_15)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails") +
ylim(15)
ggplot(df_15, aes(d, num_over_15)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails")
ggplot(df_15, aes(d, num_over_15)) +
geom_line() +
ggtitle("Number of emails over Weight 15 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails") +
ylim(c(0,15))
ggplot(df_15, aes(d, num_over_15)) +
geom_line() +
ggtitle("Number of emails over Weight 15 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails") +
ylim(c(0,550))
ggplot(df_15, aes(d, num_over_15)) +
geom_line() +
ggtitle("Number of emails over Weight 15 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails")
ggplot(df_15, aes(d, num_over_15)) +
geom_line() +
ggtitle("Number of emails over Weight 15 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails") +
ylim(c(0,550))
ggplot(df_36, aes(d, num_over_36)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails")
ggplot(df_36, aes(d, num_over_36)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails") +
ylim(c(0,550)
ggplot(df_36, aes(d, num_over_36)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails") +
ylim(c(0,550)
)
ggplot(df_36, aes(d, num_over_36)) +
geom_line() +
ggtitle("Number of emails over Weight 36 each day\nlast 2 years") +
labs(x="Day", y="Number of Emails") +
ylim(c(0,550))
summary(df_36$num_over_36)
boxplot(df_36$num_over_36)
q_got_thru = "
select
*
from
Message JOIN MessageSender USING(messageId)
JOIN MessageContent USING(messageId)
JOIN MessageSource USING(messageId)
JOIN MessageStatus USING(messageId)
WHERE
email='angela25siingh@hotmail.com' AND
DATE(creationDate)='2014-10-23'
ORDER BY
creationDate DESC
";
df_got_thru=runQuery(q_got_thru, "LeadInbox")
df_got_thru$creationDate=as.POSIXlt(df_got_thru$creationDate)
df_got_thru$spam=0
q_spam = "
select
*
from
BlockedEmail
WHERE
email_from='angela25siingh@hotmail.com'
";
df_spam=runQuery(q_spam, "Spam")
df_spam$spam=1
head(df_got_thru)
df_spam$creationDate=as.POSIXlt(df_spam$creationDate)
df_spam$creationDate=as.POSIXlt(df_spam$creationDate)
df_spam$creationDate=as.POSIXlt(df_spam$creation_date)
df_got_thru=df_got_thru[,c("creationDate", "spam")]
df_spam=df_spam[,c("creationDate", "spam")]
df_final=rbind(df_spam,df_got_thru)
ggplot(df_final, aes(creationDate, spam)) + geom_point()
head(df_final)
class(df_final$creationDate)
class(df_final$spam)
ggplot(df_final, aes(creationDate, spam)) + geom_point()
ggplot(df_final, aes(creationDate, spam))
library(ggplot2)
ggplot(df_final, aes(creationDate, spam))
ggplot(df_final)
plot(df_final)
update.packages()
remove.packages("ggplot2")
q_blocked = "
SELECT
DATE(creation_date) day,
email_type,
COUNT(*) num_emails
FROM
BlockedEmail
WHERE
DATE(creation_date) > NOW() - INTERVAL 1 YEAR
AND email_type IN (88, 55, 81, 121)
GROUP BY
1, 2
"
df_blocked=runQuery(q_blocked, "Spam")
df_blocked$day=as.POSIXct(df_blocked$day)
ggplot(df_blocked, aes(x=day, fill=as.factor(email_type)) + geom_bar()
ggplot(df_blocked, aes(x=day, fill=as.factor(email_type))) + geom_bar()
email_type
ggplot(df_blocked, aes(x=day, fill=as.factor(email_type))) + geom_bar()
source('~/.active-rstudio-document', echo=TRUE)
source('~/.active-rstudio-document', echo=TRUE)
df$day = as.POSIXlt(df$day)
ggplot(df, aes(day, num_email_addresses)) + geom_line()
source('~/.active-rstudio-document', echo=TRUE)
source('~/.active-rstudio-document', echo=TRUE)
source('~/.active-rstudio-document', echo=TRUE)
plot(df_ts)
plot(df_components)
plot(df_seasonallyadjusted);
library(choroplethr)
data(df_pop_county)
county_choropleth(df_pop_county)
county_choropleth(df_pop_county, buckets=1)
library(choroplethr)
data(df_pop_state)
choro = StateChoropleth$new(df_pop_state)
choro$show_labels=FALSE
choro$render()
?state_choropleth
data(state.names)
library(choroplethr)
data(state.names)
data(state.regions)
data(state.region)
data(state.regions)
library(choroplethrMaps)
data(state.regions)
data(state.regions)
library(choroplethrMaps)
?state.regions
data(state.regions)
head(state.regions)
update.packages()
remove.packages("choroplethr")
remove.packages("choroplethrMaps")
install.packages("choroplethr")
install.packages("choroplethr")
source('~/.active-rstudio-document', echo=TRUE)
source('~/.active-rstudio-document', echo=TRUE)
install.packages("choroplethr")
install_github("trulia/choroplethr")
library(choroplethr)
source('~/.active-rstudio-document', echo=TRUE)
install_github("trulia/choroplethrMaps")
library(choroplethrMaps)
library(choroplethr)
source('~/.active-rstudio-document', echo=TRUE)
head(test_df)
source('~/.active-rstudio-document', echo=TRUE)
source('~/.active-rstudio-document', echo=TRUE)
source('~/.active-rstudio-document', echo=TRUE)
update.packages()
library("shiny", lib.loc="/Library/Frameworks/R.framework/Versions/3.0/Resources/library")
runExample("01_hello")
runExample("02_text") # tables and data frames
runExample("03_reactivity")
runExample("04_mpg")
runExample("05_sliders")
runExample("06_tabsets")
runExample("07_widgets")
runExample("08_html")
runExample("09_upload")
shiny::runApp()
shiny::runApp('code/R/shiny')
counties <- readRDS("census-app/data/counties.rds")
getwd()
setwd("R/shiny")
getwd()
setwd("code/r/")
setwd("shiny")
counties <- readRDS("census-app/data/counties.rds")
counties <- readRDS("data/counties.rds")
head(counties)
library(maps)
library(mapproj)
install.packages("mapproj")
library(mapproj)
percent_map(counties$white, "darkgreen", "% white")
# Note: percent map is designed to work with the counties data set
# It may not work correctly with other data sets if their row order does
# not exactly match the order in which the maps package plots counties
percent_map <- function(var, color, legend.title, min = 0, max = 100) {
# generate vector of fill colors for map
shades <- colorRampPalette(c("white", color))(100)
# constrain gradient to percents that occur between min and max
var <- pmax(var, min)
var <- pmin(var, max)
percents <- as.integer(cut(var, 100,
include.lowest = TRUE, ordered = TRUE))
fills <- shades[percents]
# plot choropleth map
map("county", fill = TRUE, col = fills,
resolution = 0, lty = 0, projection = "polyconic",
myborder = 0, mar = c(0,0,0,0))
# overlay state borders
map("state", col = "white", fill = FALSE, add = TRUE,
lty = 1, lwd = 1, projection = "polyconic",
myborder = 0, mar = c(0,0,0,0))
# add a legend
inc <- (max - min) / 4
legend.text <- c(paste0(min, " % or less"),
paste0(min + inc, " %"),
paste0(min + 2 * inc, " %"),
paste0(min + 3 * inc, " %"),
paste0(max, " % or more"))
legend("bottomleft",
legend = legend.text,
fill = shades[c(1, 25, 50, 75, 100)],
title = legend.title)
}
percent_map(counties$white, "darkgreen", "% white")
shiny::runApp()
shiny::runApp()
shiny::runApp()
shiny::runApp()
runApp("stockVis")
getwd()
runApp("stockVis")
library(shiny)
runApp("stockVis")