从2D数组python中删除引号

Question

I am currently trying to execute code that evaluetes powers with big exponents without calculating them, but instead logs of them. I have a file containing 1000 lines. Each line contains two itegers separated by a comma. I got stuck at point where i tried to remove quotes from array. I tried many way of which none worked. Here is my code:

function from myLib called split() takes two argumanets of which one is a list and second is to how many elemts to split the original list. Then does so and appends smaller lists to the new one.

import math
import myLib

i = 0
record = 0
cmpr = 0
with open("base_exp.txt", "r") as f:
    fArr  = f.readlines()
    fArr  = myLib.split(fArr, 1)
    #place get rid of quotes
    print(fArr)
    while i < len(fArr):
        cmpr = int(fArr[i][1]) * math.log(int(fArr[i][0]))
        if cmpr  > record:
            record = cmpr
            print(record)
        i = i + 1

This is how my Array looks like:

[['519432,525806\n'], ['632382,518061\n'], ... ['172115,573985\n'], ['13846,725685\n']]

I tried to find a way around the 2d array and tried:

i = 0
record = 0
cmpr = 0
with open("base_exp.txt", "r") as f:
    fArr  = f.readlines()
    #fArr  = myLib.split(fArr, 1)
    fArr = [x.replace("'", '') for x in fArr]
    print(fArr)
    while i < len(fArr):
        cmpr = int(fArr[i][1]) * math.log(int(fArr[i][0]))
        if cmpr  > record:
            record = cmpr
            print(i)
        i = i + 1

But output looked like this:

['519432,525806\n', '632382,518061\n', '78864,613712\n', ...

And the numbers in their current state cannot be considered as integers or floats so this isnt working as well...:

[int(i) for i in lst]

Expected output for the array itself would look like this, so i can pick one of the numbers and work with it:

[[519432,525806], [632382,518061], [78864,613712]...

I would really apreciate your help since im still very new to python and programming in general.

Thank you for your time.

Answer 1

You can avoid all of your problems by simply using numpy 's convenient loadtxt function:

import numpy as np
arr = np.loadtxt('p099_base_exp.txt', delimiter=',')
arr

array([[519432., 525806.],
       [632382., 518061.],
       [ 78864., 613712.],
       ...,
       [325361., 545187.],
       [172115., 573985.],
       [ 13846., 725685.]])

If you need a one-dimensional array:

arr.flatten()
# array([519432., 525806., 632382., ..., 573985.,  13846., 725685.])

Answer 2

This snippet will transform your array to 1D array of integers:

from itertools import chain

arr = [['519432,525806\n'], ['632382,518061\n']]

new_arr = [int(i.strip()) for i in chain.from_iterable(i[0].split(',') for i in arr)]
print(new_arr)

Prints:

[519432, 525806, 632382, 518061]

For 2D output you can use this:

arr = [['519432,525806\n'], ['632382,518061\n']]

new_arr = [[int(i) for i in v] for v in (i[0].split(',') for i in arr)]
print(new_arr)

This prints:

[[519432, 525806], [632382, 518061]]

Answer 3

This is your missing piece:

fArr = [[int(num) for num in line.rstrip("\n").split(",")] for line in fArr] 

Here, rstrip("\\n") will remove trailing \\n character from the line and then the string will be split on , so that each string will be become a list and all integers in that line will become elements of that list but as a string. Then, we can call int() function on each list element to convert them into int data type.

Below code should do the job if you don't want to import an additional library.

i = 0
record = 0
cmpr = 0
with open("base_exp.txt", "r") as f:
    fArr = f.readlines()
    fArr = [[int(num) for num in line.rstrip("\n").split(",")] for line in fArr] 
    print(fArr)
    while i < len(fArr): 
        cmpr = fArr[i][1] * math.log(fArr[i][0])
        if cmpr > record:
            record = cmpr
            print(i) 
        i = i + 1

Answer 4

new_list=[]
a=['519432,525806\n', '632382,518061\n', '78864,613712\n',]
for i in a:
    new_list.append(list(map(int,i.split(","))))

print(new_list)

Output:

[[519432, 525806], [632382, 518061], [78864, 613712]]

In order to flatten the new_list

from functools  import reduce
reduce(lambda x,y: x+y,new_list)
print(new_list)

Output:

[519432, 525806, 632382, 518061, 78864, 613712]