CODEWITHSUNDEEP
pythonpandasdataframe
Qulymzhanov Nursat
Qulymzhanov Nursat

Reputation: 7

Filling empty column with condition

I wanna fill my dataset with condition, by using columns in dataset. You see it here

And by column "Average hour", I create column "Car type" and fill it by this function:

def sample(df, i, steps):
    for i in range(steps):
        if(df["Average hour"].all())<70:
            df["Car type"].fillna("Mini truck").all()
        elif(df["Average hour"].all()>70 and df["Average hour"].all()<90):
            df["Car type"].fillna("VAN").all()
        elif(df["Average hour"].all()>90 and df["Average hour"].all()<100):
            df["Car type"].fillna("Bus").all()
        elif(df["Average hour"].all()>100 and df["Average hour"].all()<120):
            df["Car type"].fillna("SUV").all()
        elif(df["Average hour"].all()>120):
            df["Car type"].fillna("PickUP truck").all()
    return df

When I created new column, it has full NaN values, according to this point, I used .fillna(), but terminal tells me use .all() too, but I still confused, this function isn't working. If you will advice me, write with np.where, can you explain, how I use it? May be I miss something?

Upvotes: 0

Views: 58

Answers (2)

Timeless
Timeless

Reputation: 37737

Here is another way to create a categorical column. It's by using pandas.cut.

bins = [0, 70, 90, 100, 120, float("inf")]
labels = ['Mini truck', 'VAN', 'bus', 'SUV', 'PickUP truck']

df['Car Type'] = pd.cut(df['Average hour'], bins, labels=labels)

>>> print(df)

enter image description here

Upvotes: 2

mosc9575
mosc9575

Reputation: 6337

Based on your image I think apply() with a custom function f solves it for you.

def f(x):
    if x<70:
        return "Mini truck"
    elif 70<x<90: # because of the logic befor this cloud simplified to x<90
        return "VAN"
    elif 90<x<100:
        return "Bus"
    elif 100<x<120:
        return "SUV"
    elif 120<x:
        return "PickUP truck"

df["Car type"] = df["Average Hour"].apply(f)

Minimal Example

import pandas as pd
from io import StringIO

t="""Brand  Average Hour
Audi  122
BWM  89
"""

df = pd.read_csv(StringIO(t), sep="\s\s", index_col=0)
df["Car type"] = df["Average Hour"].apply(f)

>>> df 
       Average Hour      Car type
Brand                            
Audi            122  PickUP truck
BWM              89           VAN

Upvotes: 1

Related Questions