EXTRACT_COLUMNS
Download Flojoy Studio to try this app
Take an input dataframe/matrix and returns a dataframe/matrix with only the specified columns. Params: default : DataFrame|Matrix Input to use as the table for column extraction columns : list of str or list of int The columns to extract from the input dataframe Returns: out : DataFrame|Matrix DataFrame or Matrix with only the specified columns
Python Code
import numpy as np
from flojoy import Array, DataFrame, Matrix, flojoy
@flojoy
def EXTRACT_COLUMNS(default: DataFrame | Matrix, columns: Array) -> DataFrame:
"""Take an input dataframe/matrix and returns a dataframe/matrix with only the specified columns.
Parameters
----------
default : DataFrame|Matrix
Input to use as the table for column extraction
columns : list of str or list of int
The columns to extract from the input dataframe
Returns
-------
DataFrame|Matrix
DataFrame or Matrix with only the specified columns
"""
if isinstance(default, DataFrame):
df = default.m
new_df = df[columns.unwrap()] if columns else df
return DataFrame(df=new_df)
else:
matrix = default.m
indices = np.array(columns.unwrap(), dtype=int)
new_matrix = matrix[:, indices] if columns else matrix
return Matrix(m=new_matrix)
Example App
Having problems with this example app? Join our Discord community and we will help you out!
In this example, we load tips
dataset from PLOTLY_DATASET
node and extract columns tips, total bill
and time
using two separate EXTRACT_COLUMNS
node. The results are displayed through TABLE
node