For loop in DataFrame - python-3.x

Question

I have multiples files with a lot of data and 19 columns. I am trying to to multiple for-loop and set it equal the first column, second etc. in the files.
import numpy as np
import glob
import pandas as pd
#
lat=np.zeros(90)
long=np.zeros(180)
indat=np.zeros(19)
#
file_in = glob.glob('filenames*.dat').
for a in range(140):
for i in range (90):
for j in range (180):
df = pd.DataFrame()
for f in file_in:
cols = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18] #there are nineteen columns
indat = df.append(pd.read_csv(f, delimiter='\\s+', header=None, usecols=cols, skiprows=4), ignore_index=True)
lat[i]=indat[0] # error here
long[j]=indat[1]
#updates some code here
if i >=70:
dens[a,j,i-70]=indat[2]
It gave me this error:
ValueError: setting an array element with a sequence.
Updates:
indat has 19 columns, many files but all the format is the same.
Sample indat
#columns
#0 1 2 3 ..... 19
-90 0 2e-12 #just some number
-90 2 3e-12 #just some number
-90 4 4e-12 #just some number
...
-90 360 1e-12 #just some number
-88 0 1e-11 #just some number
-88 2 2e-11 #just some number
-88 4 3e-11 #just some number
...
-88 360 4e-11 #just some number
...
90 0 2.5e-12 #just some number
90 2 3.5e-11 #just some number
90 4 4.5e-12 #just some number
...
90 360 1.5e-12 #just some number
EDIT: I clean the code up based on everyone suggestions
import numpy as np
import glob
import pandas as pd
file_in = glob.glob('filenames*.dat').
df = pd.DataFrame()
for f in file_in:
cols = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18]
indat = pd.read_csv(f, delimiter='\\s+', header=None, usecols=cols, skiprows=4)
for a in range(140):
for i in range (90):
for j in range (180):
lat[i]=indat[0] # error here
long[j]=indat[1]
if i >=70:
dens[a,j,i-70]=indat[2]

Answer 1

you tried to assign a column (pandas series) indat[0] to an element of a numpy vector lat[i]
Also what the point of indat=np.zeros(19) when you override it to be a dataframe later?
What is the content of indat[0]?
This line of code
indat = df.append(pd.read_csv(f, delimiter='\\s+', header=None, usecols=cols, skiprows=4), ignore_index=True)
is basically same as
indat = pd.read_csv(f, delimiter='\\s+', header=None, usecols=cols, skiprows=4)
because df never changed, i.e. it is always an empty dataframe
Since the content of indat is unknown, it's difficult to fix your code.
If you just want to make it run without error, I suggest to write
lat[i] = indat[0].values[0] # take the first value of the vector
long[i] = indat[1].values[0] # take the first value of the vector
It's good to take some tutorial on Numpy and Pandas since it can be very confusing without some basic understanding.