User Tools

Site Tools


other:python:jyp_steps

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
Next revision Both sides next revision
other:python:jyp_steps [2016/01/29 15:41]
jypeter Added the numpy for matlab users references
other:python:jyp_steps [2016/02/19 15:17]
jypeter Added more debugger mode instructions
Line 49: Line 49:
 Where: [[http://​docs.scipy.org/​doc/​|html and pdf documentation]] Where: [[http://​docs.scipy.org/​doc/​|html and pdf documentation]]
  
-How to get started?+==== Getting ​started ​==== 
   - always remember that indices start at ''​0''​ and that the last element of an array is at index ''​-1''​!\\ First learn about //​indexing//​ and //slicing// by manipulating strings, as shown in [[#​part1|Part 1]] above (try '''​This document by JY is awesome!'​[::​-1]''​ and '''​This document by JY is awesome!'​[slice(None,​ None, -1)]''​) 8-)   - always remember that indices start at ''​0''​ and that the last element of an array is at index ''​-1''​!\\ First learn about //​indexing//​ and //slicing// by manipulating strings, as shown in [[#​part1|Part 1]] above (try '''​This document by JY is awesome!'​[::​-1]''​ and '''​This document by JY is awesome!'​[slice(None,​ None, -1)]''​) 8-)
   - if you are a Matlab user (but the references are interesting for others as well), you can read the following:   - if you are a Matlab user (but the references are interesting for others as well), you can read the following:
Line 59: Line 60:
     - Numpy Reference Guide     - Numpy Reference Guide
     - Scipy Reference Guide     - Scipy Reference Guide
 +
 +==== Beware of the array view side effects ====
 +
 +<note warning>​When you take a slice of an array, you get a **//​View//​** : an array that has a new shape but that still shares its data with the first array.
 +
 +That is not a problem when you only read the values, but **if you change the values of the //View//, you change the values of the first array** (and vice-versa)! If that is not what want, do not forget to **make a copy** of the data before working on it!
 +
 +//Views// are a good thing most of the time, so only make a copy of your data when needed, because otherwise copying a big array will just be a waste of CPU and computer memory. Anyway, it is always better to understand what you are doing... :-P
 +
 +Check the example below and the [[https://​docs.scipy.org/​doc/​numpy-dev/​user/​quickstart.html#​copies-and-views|copies and views]] part of the quickstart tutorial.
 +
 +<code python>
 +>>>​ import numpy as np
 +>>>​ a = np.arange(30).reshape((3,​10))
 +>>>​ a
 +array([[ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9],
 +       [10, 11, 12, 13, 14, 15, 16, 17, 18, 19],
 +       [20, 21, 22, 23, 24, 25, 26, 27, 28, 29]])
 +
 +>>>​ b = a[1, :]
 +>>>​ b
 +array([10, 11, 12, 13, 14, 15, 16, 17, 18, 19])
 +
 +>>>​ b[3:7] = 0
 +>>>​ b
 +array([10, 11, 12,  0,  0,  0,  0, 17, 18, 19])
 +
 +>>>​ a
 +array([[ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9],
 +       [10, 11, 12,  0,  0,  0,  0, 17, 18, 19],
 +       [20, 21, 22, 23, 24, 25, 26, 27, 28, 29]])
 +       
 +>>>​ a[:, 2:4] = -1
 +>>>​ a
 +array([[ 0,  1, -1, -1,  4,  5,  6,  7,  8,  9],
 +       [10, 11, -1, -1,  0,  0,  0, 17, 18, 19],
 +       [20, 21, -1, -1, 24, 25, 26, 27, 28, 29]])
 +       
 +>>>​ b
 +array([10, 11, -1, -1,  0,  0,  0, 17, 18, 19])
 +
 +>>>​ c = a[1, :].copy()
 +>>>​ c
 +array([10, 11, -1, -1,  0,  0,  0, 17, 18, 19])
 +
 +>>>​ c[:] = 9
 +>>>​ c
 +array([9, 9, 9, 9, 9, 9, 9, 9, 9, 9])
 +
 +>>>​ b
 +array([10, 11, -1, -1,  0,  0,  0, 17, 18, 19])
 +
 +>>>​ a
 +array([[ 0,  1, -1, -1,  4,  5,  6,  7,  8,  9],
 +       [10, 11, -1, -1,  0,  0,  0, 17, 18, 19],
 +       [20, 21, -1, -1, 24, 25, 26, 27, 28, 29]])
 +</​code></​note>​
  
 ===== cdms2 and netCDF4 ===== ===== cdms2 and netCDF4 =====
Line 131: Line 189:
  
   * [[http://​blog.codinghorror.com/​a-pragmatic-quick-reference/​|A Pragmatic Quick Reference]]   * [[http://​blog.codinghorror.com/​a-pragmatic-quick-reference/​|A Pragmatic Quick Reference]]
 +
 +===== Debugging your code =====
 +
 +There is only so much you can do with staring at your code in your favorite text editor, and adding ''​print''​ lines in your code (or using [[https://​docs.python.org/​2/​howto/​logging.html#​logging-basic-tutorial|logging]] instead of ''​print''​). The next step is to **use the python debugger**!
 +
 +==== Debugging in text mode ====
 +
 +  - Start the script with: ''​python -m pdb my_script.py''​
 +  - Type ''​run''​ (or **r**) to go to the first line of the script
 +  - Type ''​continue''​ (or **c**) to execute the script to the end, or till the first breakpoint or error is reached
 +  - Use ''​where''​ (or **w**) to check the call stack that led to the current stop. Use ''​up''​ and ''​down''​ to navigate through the call stack and examine the values of the functions'​ parameters
 +  - Type ''​break NNN''​ to stop at line NNN
 +  - Use ''​type(var)''​ and ''​print var''​ to check the type and values of variables. You can also change the variables'​ values on the fly!
 +  - Type ''​run''​ (or **r**) to restart the script
 +  - Use ''​next''​ and ''​step''​ to execute some parts of the script line by line. If a code line calls a function:
 +    * ''​next''​ (or **n**) will execute a function and stop on the next line
 +    * ''​step''​ (or **s**) will stop at the first line **inside the function**  ​
 +  - Check the [[https://​docs.python.org/​2/​library/​pdb.html#​debugger-commands|debugger commands]] for details, or type ''​help''​ in the debugger for using the built-in help
 +
 +==== Using pydebug ====
 +
 +Depending on the distribution,​ the editor and the programming environment you use, you may have access to a graphical version of the debugger. UV-CDAT users can use ''​pydebug my_script.py''​
  
 ===== Improving the performance of your code ===== ===== Improving the performance of your code =====
other/python/jyp_steps.txt · Last modified: 2024/03/07 10:15 by jypeter