cleaning

#Clean size
df['size']=pd.to_numeric(df['size'].str.extract('(\d+)', expand=False))
df['size']+=0.5

#Clean location
df['distance_center']=pd.to_numeric(df['distance_center'].str.extract('(\d+)', expand=False))
df['distance_center'] = df['distance_center'].fillna(1)

#Clean price
new=[]
for i in range(0,len(df)):
    strings = ''.join([i for i in df.price[i] if  i.isdigit()])
    strings = strings[:-2]
    new.append(pd.to_numeric(strings))

df['price'] = new
df['price'].replace('', np.nan, inplace=True)

#Clean all and remove outliers
df=df.dropna()
df = df[df['price'] < 10000]