A5

[Ryan Sturm]

Whats a good runtime for A5?
I know mine uses a lot of memory and does a lot of calculations, so I
was wondering if there is a time I should be aiming for.

[Lonnie Princehouse]

It's not all about clock time.  If you need to spend more time computing a move, but can get a higher score, that's not necessarily bad.  That said, you'd have to be doing some heavy computing for the program to take more than a minute... the random walk example completes the goal in under two seconds on my machine (with around -2000000 points... shouldn't be hard to beat that).

--
You received this message because you are subscribed to the Google Groups "Cornell CS 2110" group.
To post to this group, send email to cornell...@googlegroups.com.
To unsubscribe from this group, send email to cornell-cs211...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/cornell-cs2110?hl=en.

[Rachel Lam]

Do you mean the RandomWalkNaturalist example code provided in the
assignment blurb? I ran it on my machine, and its been runing for
almost 5 minutes now without seeming to be nearing an end.. Am I
running the right code? Thanks!

[ninelife]

do it with --headless argument, took me 6secs (w/ -7787582 score)

--

[Kerwell]

Do we need to implement one of the graph algorithms? Currently I'm
just using logic and getting a negative score, but one that's still
much better than the example. Also, does our lowest assignment grade
get dropped?

On Nov 19, 10:55 pm, ninelife <nanlif...@gmail.com> wrote:
> do it with --headless argument, took me 6secs (w/ -7787582score)
>
>
>
> On Fri, Nov 19, 2010 at 10:44 PM, Rachel Lam <rachella...@gmail.com> wrote:
> > Do you mean the RandomWalkNaturalist example code provided in the
> > assignment blurb? I ran it on my machine, and its been runing for
> > almost 5 minutes now without seeming to be nearing an end.. Am I
> > running the right code? Thanks!
>
> > --
> >  You received this message because you are subscribed to the Google Groups
> > "Cornell CS 2110" group.
> > To post to this group, send email to cornell...@googlegroups.com.
> > To unsubscribe from this group, send email to

> > cornell-cs211...@googlegroups.com<cornell-cs2110%2Bunsubscribe@go oglegroups.com>
> > .

[Lonnie Princehouse]

There's no requirement to use the graph algorithms we've covered in class, but if you're inventing your own ad-hoc graph algorithms, you should explain them and persuade the graders that they work in your README.  

That said, it's kind of weird to avoid these algorithms. If I may use an analogy, it's like taking a carpentry class, and then saying, "A hammer?  No thanks, I'll just pound in the nails with rocks."

To unsubscribe from this group, send email to cornell-cs211...@googlegroups.com.

[Lonnie Princehouse]

And yes, we will drop the lowest assignment grade.  But there's the condition that you must actually seriously attempt the assignment.

[cfs6292]

i am trying to do little tweaks to my code to improve the runtime, but
i have been at 0.01 seconds for a while is there anywhere we can
change it to see more than two decimal places so i can see if and how
much my changes are improving my code?

On Nov 23, 12:06 pm, Lonnie Princehouse <lon...@cs.cornell.edu> wrote:
> And yes, we will drop the lowest assignment grade.  But there's the
> condition that you must actually seriously attempt the assignment.
>
> On Tue, Nov 23, 2010 at 12:05 PM, Lonnie Princehouse
> <lon...@cs.cornell.edu>wrote:
>
>
>
> > There's no requirement to use the graph algorithms we've covered in class,
> > but if you're inventing your own ad-hoc graph algorithms, you should explain
> > them and persuade the graders that they work in your README.
>
> > That said, it's kind of weird to avoid these algorithms. If I may use an
> > analogy, it's like taking a carpentry class, and then saying, "A hammer?  No
> > thanks, I'll just pound in the nails with rocks."
>

[Lonnie Princehouse]

No, but you could make a larger custom map that will take longer to run.

To unsubscribe from this group, send email to cornell-cs211...@googlegroups.com.