Code to visualize and clean outliers in Python

cleaning_outliers.py

import pandas as pd
import numpy as np

# dataviz imports
import plotly as py
import plotly.express as px
import plotly.graph_objs as go
import plotly.figure_factory as ff
from plotly.subplots import make_subplots
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot
init_notebook_mode(connected=True) 
import cufflinks as cf

# visualizing outliers
df.select_dtypes('number').quantile(np.linspace(0, 1, 1001)).iplot(logy=True,
                                                                   layout_update={'hovermode': 'x',},
                                                                   xTitle='percentile',
                                                                   yTitle='value'
                                                                   )

# replacing outliers by NaNs
def drop_outliers(data: Union[pd.DataFrame, pd.Series],
                  min_quantile=0, max_quantile=1,
                  min_value=-np.inf, max_value=np.inf):
    """Sets the values that are out of the given range set by max/min quentiles or values to NaNs
    Returns pd.DataFrame or pd.Series
    """
    result = data[
        (data > data.quantile(min_quantile))
        & (data < data.quantile(max_quantile))
        & (data > min_value)
        & (data < max_value)
    ]
    return result