Exploratory Data Analysis (EDA)

perform_eda(data, provider='pandas-profiling', sample_size=10000, data_randomizer=2)

Performs Exploratory Data Analysis (EDA)

data: pandas dataframe

Dataframe for exploratory data analysis

provider{‘pandas-profiling’, ‘sweetviz’, ‘dtale’}, default=’pandas-profiling’

Library provider for exploratory data analysis

sample_size: str, default=10000

Number of rows to return from dataframe. None to perform eda on the complete dataset which can be slower if dataset has large number of rows and columns

data_randomizer: int, default=None

Controls the data split. Provide a value to reproduce the same split.

Examples

EDA using Pandas Profiling

In [ ]:

pip install bluemist

In [ ]:

from sklearn import datasets

from bluemist.environment import initialize
from bluemist.eda import perform_eda

In [ ]:

initialize()
data = datasets.load_diabetes(as_frame=True)
perform_eda(data.frame, provider='pandas-profiling')

## Output file will be opened in the browser after analysis is completed !! ##

EDA using SweetVIZ

In [ ]:

pip install bluemist

In [ ]:

from sklearn import datasets

from bluemist.environment import initialize
from bluemist.eda import perform_eda

In [ ]:

initialize()
data = datasets.load_diabetes(as_frame=True)
perform_eda(data.frame, provider='sweetviz')

## Output file will be opened in the browser after analysis is completed !! ##

EDA using D-TALE

In [ ]:

pip install bluemist

In [ ]:

from sklearn import datasets

from bluemist.environment import initialize
from bluemist.eda import perform_eda

In [ ]:

initialize()
data = datasets.load_diabetes(as_frame=True)
perform_eda(data.frame, provider='dtale')

## dtale UI will open in the browser after analysis is completed !! ##