datatable API¶

Symbols listed here are available for import from the root of the datatable module.

Submodules¶

`exceptions.`	`datatable` warnings and exceptions.
`internal.`	Access to some internal details of `datatable` module.
`math.`	Mathematical functions, similar to python’s `math` module.
`models.`	A small set of data analysis tools.
`re.`	Functions using regular expressions.
`str.`	Functions for working with string columns.
`time.`	Functions for working with date/time columns.

Classes¶

`Frame`	Main “table of data” class. This is the equivalent of pandas’ or Julia’s `DataFrame`, R’s `data.table` or `tibble`, SQL’s `TABLE`, etc.
`FExpr`	Helper class for computing formulas over a frame.
`Namespace`	Helper class for addressing columns in a frame.
`Type`	Column’s type, similar to numpy’s `dtype`.
`stype`	[DEPRECATED] Enum of column “storage” types.
`ltype`	[DEPRECATED] Enum of column “logical” types.

Functions¶

`fread()`	Read CSV/text/XLSX/Jay/other files
`iread()`	Same as `fread()`, but read multiple files at once

`by()`	Group-by clause for use in Frame’s square-bracket selector
`join()`	Join clause for use in Frame’s square-bracket selector
`sort()`	Sort clause for use in Frame’s square-bracket selector
`update()`	Create new or update existing columns within a frame

`cbind()`	Combine frames by columns
`rbind()`	Combine frames by rows
`repeat()`	Concatenate frame by rows

`as_type()`	Cast column into another type
`categories()`	Get categories for categorical columns
`ifelse()`	Ternary if operator
`shift()`	Shift column by a given number of rows
`cut()`	Bin a column into equal-width intervals
`qcut()`	Bin a column into equal-population intervals
`split_into_nhot()`	[DEPRECATED] Split and nhot-encode a single-column frame

`init_styles()`	Inject datatable’s stylesheets into the Jupyter notebook
`rowall()`	Row-wise all() function
`rowany()`	Row-wise any() function
`rowcount()`	Calculate the number of non-missing values per row
`rowfirst()`	Find the first non-missing value row-wise
`rowlast()`	Find the last non-missing value row-wise
`rowargmax()`	Find the index of the largest element row-wise
`rowmax()`	Find the largest element row-wise
`rowmean()`	Calculate the mean value row-wise
`rowargmin()`	Find the index of the smallest element row-wise
`rowmin()`	Find the smallest element row-wise
`rowsd()`	Calculate the standard deviation row-wise
`rowsum()`	Calculate the sum of all values row-wise

`intersect()`	Calculate the set intersection of values in the frames
`setdiff()`	Calculate the set difference between the frames
`symdiff()`	Calculate the symmetric difference between the sets of values in the frames
`union()`	Calculate the union of values in the frames
`unique()`	Find unique values in a frame

`corr()`	Calculate correlation between two columns
`count()`	Count non-missing values per column
`countna()`	Count the number of NA values per column
`cumcount()`	Number rows within each group
`cummax()`	Calculate the cumulative maximum of values per column
`cummin()`	Calculate the cumulative minimum of values per column
`cumprod()`	Calculate the cumulative product of values per column
`cumsum()`	Calculate the cumulative sum of values per column
`cov()`	Calculate covariance between two columns
`fillna()`	Impute missing values
`max()`	Find the largest element per column
`mean()`	Calculate mean value per column
`median()`	Find the median element per column
`min()`	Find the smallest element per column
`ngroup()`	Number each group
`nunique()`	Count the number of unique values per column
`prod()`	Calculate the product of all values per column
`sd()`	Calculate the standard deviation per column
`sum()`	Calculate the sum of all values per column

Other¶

`build_info`	Information about the build of the `datatable` module.
`dt`	The `datatable` module itself.
`f`	The primary namespace used during `DT[...]` call.
`g`	Secondary namespace used during `DT[..., join()]` call.
`options`	`datatable` options.