I am trying to pass a column of numbers to a Python script to then convert into a numpy array.
input.txt
42
42.4
43.5153
44
Bash Code
python script.py ${input}
Python Script
#!/usr/bin/env python3
import sys
import numpy
in = sys.argv[1]
in_out = np.array([float(in)])
print "Inputs:" in_out
sys.exit()
Python Error
Traceback (most recent call last):
File "script.py", line 8, in <module>
in_out = np.array([float(in)])
ValueError: invalid literal for float(): 42
import numpy as np
with open('input.txt') as input_file:
data = np.array([float(line.strip()) for line in input_file])
You need to cast all values as float so that numpy is only storing one datatype in the array.
If you want to supply the file as an argument, you can do the following:
import numpy as np
import sys
file_name = sys.argv[1]
with open(file_name) as input_file:
data = np.array([float(line.strip()) for line in input_file])
I think the problem is that your in variable (which can't be a variable in Python, but we'll go with that) is a string of all values. So you need to split it and apply float to each individually:
np.array(list(map(float, in.split())))
But I'd suggest reading an input.txt file directly from Python.
Use np.loadtxt
Ex:
import numpy as np
in_out = np.loadtxt(filename, dtype=float)
print(in_out)
Output:
[42. 42.4 43.5153 44. ]
in_inand${input}? I presume you need to read the file instead of reading and parsing the filename as a float