如何拆分标题值并将它们转换为不同的列？

Question

I have a data which looks like this

However I need to separate the header values and convert them into three different column and it should look like the following.

What is the best method to achieve this?

import matplotlib.pyplot as plt
import pandas as pd
import math
import os

url = 'https://raw.githubusercontent.com/callmedemi/AMPSE/main/machinepara1_gm_m1.csv'
dataset = pd.read_csv(url, encoding='latin')
print(dataset)
dataset1 = dataset.columns
print(dataset1)

The answer looks like this:

        wn   ...  leafValue( OP("/I0/M1" "gm") "Vbias" 2 "lp" 4e-05 "wp" 0.0002 ) (S)
0  0.000005  ...                                       6.773000e-21                  
1  0.000008  ...                                       6.774000e-21                  
2  0.000010  ...                                       6.775000e-21                  
3  0.000013  ...                                       6.776000e-21                  
4  0.000015  ...                                       6.776000e-21                  

[5 rows x 251 columns]
Index(['wn ',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 0 "lp" 1.5e-05 "wp" 0.00012 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 0 "lp" 1.5e-05 "wp" 0.00014 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 0 "lp" 1.5e-05 "wp" 0.00016 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 0 "lp" 1.5e-05 "wp" 0.00018 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 0 "lp" 1.5e-05 "wp" 0.0002 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 0 "lp" 2.125e-05 "wp" 0.00012 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 0 "lp" 2.125e-05 "wp" 0.00014 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 0 "lp" 2.125e-05 "wp" 0.00016 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 0 "lp" 2.125e-05 "wp" 0.00018 ) (S)',
       ...
       'leafValue( OP("/I0/M1" "gm") "Vbias" 2 "lp" 3.375e-05 "wp" 0.00012 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 2 "lp" 3.375e-05 "wp" 0.00014 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 2 "lp" 3.375e-05 "wp" 0.00016 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 2 "lp" 3.375e-05 "wp" 0.00018 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 2 "lp" 3.375e-05 "wp" 0.0002 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 2 "lp" 4e-05 "wp" 0.00012 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 2 "lp" 4e-05 "wp" 0.00014 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 2 "lp" 4e-05 "wp" 0.00016 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 2 "lp" 4e-05 "wp" 0.00018 ) (S)',
       'leafValue( OP("/I0/M1" "gm") "Vbias" 2 "lp" 4e-05 "wp" 0.0002 ) (S)'],
      dtype='object', length=251)

If you see the for each "lp", "wp" gives five different values and for each "Vbias"(range:0-2) there are five different lp values. The header is the main issue here as I can't extract the header.

What I want to do is, extract "Vbias", "lp" and "wp" from the header for each of "gm" values and place them in corresponding columns. The picture files are just for a better understanding, the real CSV file is in the code.

Answer 1

When trying to read the CSV file from the provided URL, I got and encoding error, so I first needed to an argument to specify a valid encoding to be used.

Then, since the column headers seem to be unusable and we know the exact names we want to put there, we may simply replace the data frame columns names with our own list:

import pandas as pd

url = 'https://raw.githubusercontent.com/callmedemi/AMPSE/main/trial.csv'
dataset = pd.read_csv(url, encoding='latin')

# remove imported column headers
df = dataset[1:]  

# and define new ones, based on our knowledge of the data
df.columns = ['wm', 'gm', 'Vbias', 'lp', 'wp' ] 

print(df)

Result:

         wm            gm         Vbias            lp            wp
1  0.000007  4.250000e-08  4.350000e-08  4.440000e-08  4.540000e-08
2  0.000010  5.480000e-08  5.580000e-08  5.670000e-08  5.770000e-08
3  0.000013  6.710000e-08  6.810000e-08  6.900000e-08  7.000000e-08
4  0.000015  7.940000e-08  8.040000e-08  8.130000e-08  8.230000e-08