# read column by indexing
# classify each column as either discrete or continuous
cols = pd.read_csv('name.csv').columns 
read_col_indexing = pd.read_csv('name.csv',usecols=cols[:5])
read_col_indexing.head()