Notes on R

R_notes.R

#### write a dataframe as a csv
write.table(df, file="<filename>", row.names=F, sep=",",  quote=F)

#### select dataframe rows by column value
df[df$state=="<colname>",]

#### select columns
# by column index
want_cols <- c(7, 8, 12)
trimmed_df <- df[,want_cols]

#### ordering/sorting dataframe based on a column(s)
df[order(df$colname),]

# reversed
df[order(-df$colname),]

#### Histograms
# breaks
hist(df$chucks, breaks=seq(0, 1, by=.001))

# ignore outliers with xlim/ylim
hist(df$chucks, breaks=seq(0, 1, by=.001), xlim=c(0, .25))

# count unique rows in a column
df.chuck_counts = as.data.frame(table(df$chucks))

# find duplicates: http://stackoverflow.com/questions/12495345/find-indices-of-duplicated-rows
dups = duplicated(df) | duplicated(df, fromLast=T)
df[dups,]

# summary statistics of a single column
unique_chucks = as.data.frame(as.table(summary(df$chucks)))