Python

Search

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)]

Comment

PREVIOUS	NEXT

Code Example
Python :: append 1 colimn in pandas df
Python :: raise_for_status() requests
Python :: python loop index and value
Python :: csv read python
Python :: python create random mac
Python :: how to print during multiprocessing
Python :: how to convert string to integer in python
Python :: matrix diagonal sum leetcode in Python
Python :: append in python
Python :: pytorch squeeze
Python :: django email change sender name
Python :: argparse accept only few options
Python :: torch print full tensor
Python :: binary to decimal in python without inbuilt function
Python :: python boolean operators
Python :: python curses for windows
Python :: python A string float numeral into integer
Python :: python web framework
Python :: django from
Python :: python check if key exist in json
Python :: defaultdict python
Python :: brute force string matching algorithm in python
Python :: get first element of array python
Python :: from django.core.management import execute_from_command_line ImportError: No module named django.core.management
Python :: Get more than one longest word in a list python
Python :: pass 2d array to 1d python
Python :: for loop with index python
Python :: arange float step
Python :: while not command in python
Python :: Progress Bars in Python

Search

PYTHON

scikit learn to identify highly correlated features

ADD CONTENT