darkside wrote:
> Maybe because of the \n? Do I have to put them off before trying fromstring?
yes == fromstring/fromfile doesn't deal with mixing \n and space (or any
other sep char), which is too bad, because this is a common use.
They could use some work in general.
but if you do a: s = s.re
Thank your, but:
If the initial list is:
>>> a
' 0.0E+00 1.806088833E-02-4.959341557E+07 0.0E+00
0.0E+00\n'
y = np.fromstring(a2,sep=' ')
>>> y
array([ 0.e+00, 1.80608883e-02, -4.95934156e+07,
0.e+00, 0.e+00])
It works perfectly.
But
darkside writes:
x = np.fromstring(b,sep='')
That should be sep=' ' (a space) to make it read ASCII representations
of numbers. Otherwise it will interpret the string as binary, and if the
string length happens to be a multiple of 8, you get garbage data,
otherwise it raises the following e
Thanks, your solution is very good.
But I have still one problem: the last line makes this:
>>> x = np.fromstring(b,sep='')
>>> x
array([ 8.56506822e-072, 2.92021791e-033, 2.25187638e+006, ...,
6.99021630e-077, 1.39804329e-076, 2.66133772e-260])
As you can see, it doesn't load t
How about
f = open(file)
s = f.read()
f.close()
a = s.replace('E-','EE').replace('-',' -').replace('EE','E-')
x = np.fromstring(a, sep=' ')
Gary R.
darkside wrote:
> I think that not all de numbers use the same number of characters, the
> problem is the signed ones. This numbers use one more ch
I think that not all de numbers use the same number of characters, the
problem is the signed ones. This numbers use one more character for '-' and
join the previous column.
Your idea is quite good, but I don't know how to do it:
If I use the pylab.load code, I suppose that I have to put something i
Hi,
I found a problem loading data: I have a file of signed numbers like:
-1.370674456E+02-1.662854139E+02 0.0E+00 0.0E+00
0.0E+00
6.964865599E+10 8.394798002E-11 4.348192658E+03 9.455873862E+02
3.292484968E-09
The problem is that the signed numbers join so I always find
