在python中分类和恢复数据

Question

I have a dataset in which resides in a 13 by 506 matrix, let's call the data set data_1. I am interested in one of the columns data, lets call that data column data_c1. Data_c1 is numeric, so the 50th percentile can be calculated with the numpy library.

My goal is to go through data_c1, do a binary classification on whether it is above or below the 50th percentile (y=1 for above, y=0 for below) and store that information in a new matrix with the corresponding tag (y=1 or y=0.)

I figured out how to load the data and calculate t50 (see below.) Can someone show me how to complete the reclassification? I think I would need to use a while loop, but I can't get it to restore the data into a new matrix.

Here is my code so far:

#import libraries
import numpy as np
import pandas as pd

#import data set
from datasoure import data_file
data_file = data_1()
data_1['data_c1'] = data_c1

#calculate percentile using numpy
t50 = np.percentile(data_1, 50)

#classify target data as y=1 for >=t50 or <=t50
#while loop????

Answer 1

You can apply a function like this:

def classifier(row):
         global t50 #defined somewhere else
         if row["data_c1"] > t50:
                 return 1
         else:
                 return 0
     
 new_col = df.apply(classifier, axis=1)

Then you can do whatever you want with new_col