UC Berkeley CS188 Intro to AI

Frequently asked questions

Computing environments

• Can I complete the projects at home?
  

  We will only officially support the use of the instructional computing facilities. This is also a recommended method for several reasons: the codebase is easy to access and update; you will likely encounter other students at work on the same problem -- potential partners; and ultimately, assignment submission must be from your instructional account.

  However, you may prefer to work from home. The simplest option is to remotely login to an instructional machine using an SSH client (for Windows: PuTTY or Host Explorer). Some login servers you can use are {cory.eecs, quasar.eecs, star.cs, solar.cs}.berkeley.edu. The main problem with remote login is that editing code and debugging can be a pain within a terminal window. To display graphics and windows, you could install an X windowing system (for Windows: Hummingbird Exceed or XFree86). However, depending on your Internet connection, the graphics could transmit slowly -- resulting in an excruciating keystroke lag.

  To develop your code at home, you must install Python on your system. You can download this freely at www.python.org. These versions often come packaged with IDLE, a minimalist Integrated Development Environment that should be sufficient for editing and testing Python code. Alternatively, you could use the more complex but helpful IDE Eclipse IDE (Eclipse tutorial). Or, you could use Cygwin to emulate a Linux terminal on Windows. There are Cygwin packages available for many *nix tools, such as Python and Emacs. Regardless of how you choose to work at home, however, you will always have to eventually transfer your files to your instructional account, where you will probably want to test them again prior to submitting.
  
  

• How do I set up Eclipse to edit Python at home?
  

  Follow the instructions in our new Eclipse tutorial. Eclipse is an integrated development environment (IDE) that will assist you in writing and debugging your python code.
  
  

• How do I set up Eclipse to edit Python in the labs?
  

  In the lab machines, Eclipse, PyDev, and Python are all installed already (we are looking into getting the extensions installed). However, you must go to the Window > Preferences > PyDev > Interpreter - Python configuration menu and add the path to the python executable using the top right New... button. You can get this path by typing which python from the command line in a terminal.
  
  

Python

• What version of Python should I use?

Versions 2.5 and 2.6 will both work for this course. The version that is guaranteed to be most compatible with the autograder and instructional machines is Python 2.6.6 (note that this is NOT the latest version, or even the latest version of the 2.x series, or even the latest version of the 2.6.x series!). Python 2.6.7 will work as well, but does not have a Windows installer. Python 2.7.2 is likely also okay, but using its 2.7.x features in your code may cause problems for the autograder. Do not use Python 3, as it is not backwards compatible and will not work with the autograder.

• I wrote a function in filename.py. How do I execute it from the Python interpreter?
  

  If you start Python from the same directory that filename.py is in (or if it's in the Python path...), then you can import the contents of filename.py as a module:

  >>> import filename
>>> filename.function(arguments)

  With this kind of import statement, you must reference the module name followed by the function name. Alternatively, you can import the contents of filename.py into the local namespace:

  >>> from filename import *
>>> function(arguments)

  With this kind of import statement, you no longer need to reference the module name.
  
  

• Why are dictionary keys in such a strange order? Should I worry about the order?
  

  Dictionaries in Python (like HashMaps in Java) are implemented using hash tables. The order of the keys depends on how exactly the hashing algorithm maps keys to buckets, and will usually seem arbitrary. Your code should not rely on key ordering, and you should not be surprised if even a small modification to how your code uses a dictionary results in a new key ordering.
  
  

• How can I enable tab completion and command history in the interactive python shell?

  One option is to install IPython, which is available for Windows as well as Linux, Unix, and Mac OS X.

  In Unix environments (Linux, Solaris, Mac OS X, Cygwin), this functionality is also available directly in Python through the readline and rlcompleter modules. To enable it, copy the following python script into a file named .pythonstartup in your home directory:

  import readline, rlcompleter

# Include this next line to enable command completion
readline.parse_and_bind("tab: complete")

# Include the following lines to enable peristent command history
# (saved to ~/.python_history)
import os, atexit
readline_history_file = os.path.join(os.path.expanduser('~'), '.python_history')
try: readline.read_history_file(readline_history_file)
except IOError: pass
readline.set_history_length(3000)
atexit.register(readline.write_history_file, readline_history_file)

  Then set the PYTHONSTARTUP environment variable to ~/.pythonstartup. If you are using the tcsh shell (the default on the inst machines), add the following line to the end of your ~/.cshrc file:

  setenv PYTHONSTARTUP ~/.pythonstartup

  If you are using bash, instead add the following to the end of your ~/.bashrc file:

  export PYTHONSTARTUP=~/.pythonstartup

---