Hi again, sorry if that wasnt clear. I am using the ipython interpreter
to debug scripts. I have a function:-) saved as part of a module called
daves_modules.py.
import numpy as np
def find_nearest(array,value):
idx=(np.abs(array-value)).argmin()
return array[idx], idx
If I run the module from the interpreter as follows,
In [610]: %run daves_modules.py
In [611]: a = find_nearest(pos_time, highTime)
In [612]: a
Out[612]: (20.009998091697867, 200)
it works fine. However, if I use the same function in a script like this,
import numpy as np
import pylab as py
from daves_modules import dist, find_nearest
.
.
.
a = find_nearest(pos_time, highTime)
I get the following error,
%run SNR_UFAN_UCRUI.py
---------------------------------------------------------------------------
IndexError Traceback (most recent
call last)
/home/davcra/python_scripts/SNR_UFAN_UCRUI.py in <module>()
88 b = find_nearest(pos_time, lowTime)
89
---> 90 pos_noise =
(np.sum(pos_signal[0:a[1]])+np.sum(pos_signal[b[1]:-1])) / (a[1] +
(len(pos_signal)-b[1]))
91 neg_noise =
(np.sum(neg_signal[0:-a[1]])+np.sum(neg_signal[-b[1]:-1])) / (a[1] +
(len(neg_signal)-b[1]))
92
IndexError: invalid index to scalar variable.
WARNING: Failure executing file: <SNR_UFAN_UCRUI.py>
If I then try to use the function in ipython again,
In [614]: a = find_nearest(pos_time, highTime)
In [615]: a
Out[615]: 20.009998091697867
Note: the function originally only returned array[idx], so it seems to
have reverted to this somehow.
On 02/24/2012 04:28 PM, Evert Rol wrote:
Hi David,
Hi,
I am new to python and have made a couple of definitions. I imported them and they worked ok. I
they worked except for one which gave me the error "NameError: global name 'np' is not
defined". I then edited my script for the def to include "import numpy as np" saved
it and imported the def again. However, it still gives me the same error. I know I have to be doing
something basic wrong but cant figure it out, anyone know what I am doing wrong. The def is below.
thanks
Minor thing first: in Python terminology, most of the time your 'definitions' are simply
called functions, although you're correct that "def" refers to definition. But
thatt's more about where the function is defined, in contrast to where in the code it is
called (or perhaps even declared, though I don't think that applies to Python).
When you say "I imported the def again", it sounds like you're working on the
Python interpreter, and doing
import myscript
or similar.
If that's how you run things, you would have to use the reload() function to
reload the new function definition, which has your correction.
However, you also talk about 'my script'. A script is something I would run
from the shell command line, like
$> python myscript.py
If you do things that way, you would always be ensured that python uses the
latest edits in your script.
It does mean that any command you would normally type into the Python
interpreter, you would now have to enter in the script. And while the
interpreter always shows the evaluation result of the last command entered, a
script would require a print for that. Compare:
a=1
a
1
versus (inside a script):
a = 1
a # this doesn't show anything
print a # this does
Perhaps this is too basic, but I have to guess a bit what you are doing from
your text.
A few tips to get more practical help:
- Python normally shows a stack trace when there is an error. It is good to
copy-paste the whole thing in your emails. Just typing the last bit often
doesn't help.
- Copy-paste (again, don't type) whatever you're doing in the Python
interpreter, if that's what you are using. So we can how you do things
(examples are clearer than descriptions). If needs be, intersperse with
comments.
Compare eg:
import myscript
NameError: global name 'np' is not defined".
# editing myscript.py
import myscript
NameError: global name 'np' is not defined".
And we can immediately see you don't reload() the script.
Hope this gets you further.
Have fun,
Evert
D
import numpy as np
def find_nearest(array,value):
idx=(np.abs(array-value)).argmin()
return array[idx]
_______________________________________________
Tutor maillist - [email protected]
To unsubscribe or change subscription options:
http://mail.python.org/mailman/listinfo/tutor
_______________________________________________
Tutor maillist - [email protected]
To unsubscribe or change subscription options:
http://mail.python.org/mailman/listinfo/tutor
