Table of Contents

dataframe

dataframe is a 2d columnar data structure that provides many powerful analysis and manipulation tools, similar to a spreadsheet or SQL engine

Functions

DataFrame

DataFrame(data, index, columns, dtype) DataFrame

constructs a DataFrame containing the given data

parameters:

name	type	description
`data`	`any`	data for the content of the DataFrame. Can be a list, dict, Series, or another DataFrame
`index`	`Index`	an Index that describes the rows
`columns`	`Index`	an Index that describes the columns
`dtype`	`string`	data type to force. If not provided, it will be inferred for each column

examples:

construct

constuct a DataFrame using a list of lists with column names

load("dataframe.star", "dataframe")
# create a new DataFrame
df = dataframe.DataFrame([["cat", "meow"],
["dog", "bark"],
["eel", "zap"]],
columns=["name", "sound"])

parse_csv

parse_csv(text) DataFrame

constructs a DataFrame by parsing the text as csv data. Assumes the first row is a header row

parameters:

name	type	description
`text`	`string`	the string to parse as csv data

Index

Index(data, name) Index

constructs an Index, which describes a single axis of a dataframe

parameters:

name	type	description
`data`	`list(string)`	a list of strings for the index
`name`	`string`	the name of the Index

Series

Series(data, index, dtype, name) Series

constructs an Series, a homogeneously typed dataframe column

parameters:

name	type	description
`data`	`list`	a list of data values. They will be coerced to use a single data type
`index`	`Index`	the index that describes the elements in the Series
`dtype`	`string`	data type of the values in the Series
`name`	`string`	name of the Series

Types

DataFrame

a dataframe

Fields

name	type	description
at	AtIndexer	returns an AtIndexer, which can be used to retrieve an arbitrary cell from the DataFrame
columns	Index	returns the columns of the DataFrame as an Index
index	Index	returns the Index of the DataFrame, if it exists
shape	tuple(int,int)	returns a tuple with the size of the DataFrame, as (number rows, number columns)

Methods

append

append(other) DataFrame

appends data to the rows of this DataFrame, returned as a new DataFrame

parameters:

name	type	description
`other`	`list`	data to append

apply

apply(function, axis) Series

travel the given axis and apply the function to each slice. The result values of that function are collected into a Series, which is returned

parameters:

name	type	description
`function`	`function`	the function to apply to each slice
`axis`	`int`	which to travel, either 0 for columns, or 1 for rows

drop

drop(labels, axis, index, columns)

drop columns or rows from the DataFrame

parameters:

name	type	description
`labels`	`list(string)`	what to drop from the DataFrame, axis is required to specify what the labels mean. axis=0 if the labels are for the index, axis=1 if the labels are for the columns
`axis`	`int`	which axis to drop from. axis=0 for index, axis=1 for columns
`index`	`list(string)`	values to drop from the index
`columns`	`Index`	values to drop from the columns

drop_duplicates

drop_duplicates(subset)

drop duplicate rows of the DataFrame

parameters:

name	type	description
`subset`	`list(string)`	which subset of each row to consider for uniqueness

groupby

groupby(by) GroupByResult

group a set of row according to some given column value

parameters:

name	type	description
`by`	`list(string)`	a list of column names to use for grouping the rows together

examples:

groupby

group rows according to the values in the given column

load("dataframe.star", "dataframe")
df = dataframe.DataFrame([["cat", "tabby"],
["cat", "black"],
["cat", "calico"],
["dog", "doberman"],
["dog", "pug"]],
columns=["species", "breed"])
num_breeds = df.groupby(['species'])['breed'].count()

head

head(n?) DataFrame

return the first n row of the DataFrame

parameters:

name	type	description
`n`	`int`	number of rows to include, defaulting to 5

merge

merge(right, left_on, right_on, how, suffixes) DataFrame

merge this with the right DataFrame, returned as a new DataFrame

parameters:

name	type	description
`right`	`DataFrame`	the DataFrame to merge with this one
`left_on`	`string`	which column of the left DataFrame to merge on
`right_on`	`string`	which column of the right DataFrame to merge on
`how`	`string`	how to merge the columns, only "inner" is supported, and is the default
`suffixes`	`list(string)`	suffixes to use for merged column names, defaulting to ["_x", "_y"]

reset_index

reset_index()

resets the index to be an empty index, turning the previous index into its own column

sort_values

sort_values(by, ascending?) DataFrame

sort the values in the DataFrame

parameters:

name	type	description
`by`	`list(string)`	the columns to use to sort the values
`ascending`	`bool`	whether to use ascending order, default is True

examples:

sort_values

sort the values

load("dataframe.star", "dataframe")
df = dataframe.DataFrame(columns=['id','animal','sound'],
data=[[1,'cat','meow'],
[2,'dog','bark'],
[3,'eel','zap'],
[4,'frog','ribbit']])
sorted = df.sort_values(by=['sound'])