/Python/Scikit learn to identify highly correlated features

Scikit learn to identify highly correlated features

# Create correlation matrix
corr_matrix = df.corr().abs()

# Select upper triangle of correlation matrix
upper = corr_matrix.where(np.triu(np.ones(corr_matrix.shape), k=1).astype(np.bool))

# Find index of feature columns with correlation greater than 0.95
to_drop = [column for column in upper.columns if any(upper[column] > 0.95)]

Source: chrisalbon.com

How to print correlation to a feature in pyhton

df[df.columns[1:]].corr()['LoanAmount'][:]

Source: datascience.stackexchange.com

Similar codes

12 votes156 views6 answers
Python convert date to timestamp
Python
0 vote129 views1 answer
How to make a dice program in python
Python
0 vote125 views1 answer
Mak a scipy csr sparse matrix
Python
0 vote133 views1 answer
Xlsb file in pandas
Python
0 vote129 views1 answer
Pytest fixtures scope explained
Python
0 vote92 views1 answer
How to get started with flask
Python
3 votes129 views2 answers
How to format a file in python
Python
0 vote128 views2 answers
Python creare una list comprehension
Python
0 vote128 views1 answer
Contextlib closing python file
Python
0 vote130 views1 answer
Pandas not a time nat
Python