Revision history [back]

The following techniques can help to make data.table subsetting faster when there are multiple criteria for exclusion:

1. Use the setkey() function to sort the data.table by the columns that will be used for subsetting. This can speed up the subsetting process as data.table can use binary search to quickly find the rows that meet the criteria.

2. Use the with = FALSE argument when subsetting with multiple conditions. This argument prevents data.table from creating a copy of the data when subsetting, which can significantly reduce memory usage and improve speed.

3. Use the %in% operator for subsetting with multiple values. The %in% operator allows you to specify multiple values in a single statement and can be faster than using multiple == operators.

4. Avoid using the which() function. The which() function can be slow when used with large data sets. Instead, use the [] operator with the ! operator to exclude rows that meet the criteria.

5. Use the frank() function to rank rows by a column before subsetting. This can be useful when subsetting based on the top or bottom n rows of a data set.

6. Use the set() function to update values in a data.table. This can be faster than using the [] operator.