Converting Datatypes

In [13]:
import pandas as pd

df = pd.read_csv('amazon_fires.csv', encoding = "ISO-8859-1")

new_column_name = { 'ano' : 'year' ,  'mes': 'month', 'estado': 'state',
                   'numero': 'fire_numbers','encontro': 'date'}
                                                
df.rename(columns= new_column_name , inplace=True)    

new_column_order = [4,1,0,2,3,]

df = df[df.columns[new_column_order]]

df['fire_numbers'] = df['fire_numbers'].str.strip(" Fires")

df.head()
Out[13]:
date month year state fire_numbers
0 1/1/1998 Janeiro 1998 Acre 0
1 1/1/1999 Janeiro 1999 Acre 0
2 1/1/2000 Janeiro 2000 Acre 0
3 1/1/2001 Janeiro 2001 Acre 0
4 1/1/2002 Janeiro 2002 Acre 0

Checking Datatypes

In [14]:
df.dtypes
Out[14]:
date            object
month           object
year             int64
state           object
fire_numbers    object
dtype: object

Converting Datatype 'fire_numbers' from object to float

In [15]:
df['fire_numbers'] = df['fire_numbers'].astype(float)

df.head()
Out[15]:
date month year state fire_numbers
0 1/1/1998 Janeiro 1998 Acre 0.0
1 1/1/1999 Janeiro 1999 Acre 0.0
2 1/1/2000 Janeiro 2000 Acre 0.0
3 1/1/2001 Janeiro 2001 Acre 0.0
4 1/1/2002 Janeiro 2002 Acre 0.0

Checking if datatype changed

In [12]:
df.dtypes
Out[12]:
date             object
month            object
year              int64
state            object
fire_numbers    float64
dtype: object