Converting exponential to float

Question

This is my code, trying to convert the second field of the line from exponential into float.

outputrrd = processrrd.communicate()
(output, error) = outputrrd
output_lines = output.split('\n')
for line in output_lines:
    m = re.search(r"(.*): ", line)
    if m != None:
        felder = line.split(': ')
        epoch =  felder[0].strip(':')
        utc = epoch2normal(epoch).strip("\n")
        #print felder[1]
        data = float(felder[1])
        float_data = data * 10000000
        print float_data
        resultslist.append( utc + ' ' + hostname + ' ' +  float_data)

But, the program stops with this error:

File "/opt/omd/scripts/python/livestatus/rrdfetch-convert.py", line 156, in <module>
    data = float(felder[1])
ValueError: invalid literal for float(): 6,0865000000e-01

Does anyone know the reason?

Your float literal contains a comma `","` instead of a decimal point `"."`. — Sven Marnach, Feb 08 '12 at 15:01

score 24 · Accepted Answer · edited Feb 08 '12 at 15:26

24

The easy way is replace! One simple example:

value=str('6,0865000000e-01')
value2=value.replace(',', '.')
float(value2)
0.60865000000000002

edited Feb 08 '12 at 15:26

juliomalegria

24,229
14
73
89

answered Feb 08 '12 at 15:17

ederwander

3,410
1
18
23

This does not change anything – alper Jan 10 '21 at 12:27
@Zorro try `print( "%.16f" % float('6.4e-05'))` – ederwander Jul 06 '21 at 15:17
thanks working, but what if i need `0.0000640` in float number without exponential and not string – Zorro Jul 06 '21 at 15:22
[site](https://www.kite.com/python/answers/how-to-suppress-scientific-notation-in-python) @Zorro – ederwander Jul 06 '21 at 15:33

NPE · Answer 2 · 2012-02-08T15:09:58.890

11

The reason is the use of comma in 6,0865000000e-01. This won't work because float() is not locale-aware. See PEP 331 for details.

Try locale.atof(), or replace the comma with a dot.

edited Feb 08 '12 at 15:09

answered Feb 08 '12 at 15:01

NPE

486,780
108
951
1,012

Pedro Lobito · Answer 3 · 2020-07-31T01:02:34.153

9

The float is correct, just use format to display it as you want, i.e.:

print(format(the_float, '.8f'))

edited Jul 31 '20 at 01:02

answered Jun 17 '17 at 10:08

Pedro Lobito

94,083
31
258
268

prem · Answer 4 · 2019-01-24T12:31:09.853

3

This work for me, try it out.

def remove_exponent(value):
    decial = value.split('e')
    ret_val = format(((float(decial[0]))*(10**int(decial[1]))), '.8f')
    return ret_val

edited Jan 24 '19 at 12:31

answered Jan 24 '19 at 12:24

prem

39
2

score 2 · Answer 5 · answered Dec 24 '13 at 11:46

I think it is useful to you:

def remove_exponent(value):
    """
       >>>(Decimal('5E+3'))
       Decimal('5000.00000000')
    """
    decimal_places = 8
    max_digits = 16

    if isinstance(value, decimal.Decimal):
        context = decimal.getcontext().copy()
        context.prec = max_digits
        return "{0:f}".format(value.quantize(decimal.Decimal(".1") ** decimal_places, context=context))
    else:
        return "%.*f" % (decimal_places, value)

score 2 · Answer 6 · answered Mar 24 '16 at 10:19

2

Simply by casting string into float:

new_val = float('9.81E7')

answered Mar 24 '16 at 10:19

0x3bfc

2,715
1
16
20

score 0 · Answer 7 · answered Oct 19 '21 at 09:30

I had a similar issue trying to convert from string in scientific/exponential notation to a float number (that can result also in exponential notation if too long)

num = '-8e-05'

def scientific_to_float(exponential):
  split_word = 'e'
  e_index = exponential.index('e')
  base = float(exponential[:e_index])
  exponent = float(exponential[e_index + 1:])
  float_number = base * (10 ** exponent)
  return float_number

scientific_to_float(num) # return -8e-05 float number

Converting exponential to float

7 Answers7

Linked