Setting out to learn Python. Anyone wanna join?

You might have a little longer, keep trying!

I got 100% 10 minutes late. I stupidly transposed an assignment and wasn't finding the max option. I'm not holding my breath; I guess I can be satisfied with a 90% since I waited until the last minute to start on it.

What's on the line with the grades?

muraii wrote:

What's on the line with the grades?

Nothing yet. They want to keep me in suspense. Or convince me that I screwed myself entirely by submitting again after the deadline and now I get ZERO.

You'll get at least the 90 you already got. That much is safe.

It might be good enough, you might have made it. Let's see!

edit: since I was focusing on this so hard for the last twenty minutes or so, I think I hit on a better strategy.

What I was doing was looking for the most efficient thing, and then checking to see if something else would let me buy that thing faster, picking the fastest one, and then recursively looking to see if I could make that purchase faster as well.

What I realize now is this: my original goal was bad. I shouldn't have been looking for the most efficient thing. What I should have done was set a goal of, hmm, probably 10 times as many cookies as I had right then, and then look for purchases that would let me get to that total the fastest. Same basic logic, just with the goal of 'more cookies', instead of 'most efficient thing'.

Yeah, the View Feedback button lists both submissions with "attaboys" whereas the 2048 submission I did for the hell of it has nothing. I wish I had bothered to submit what I had done in 2048 for at least partial credit.

Well, you didn't know it would be so long until the next course. You can most likely still pull out a 'with distinction'.

I mean, what do the grades achieve? If you're learning it it doesn't matter when you submit, except that you don't get to participate in peer review (per my Coursera course).

I want all the points!


All my With Distinction certificates will look good hanging on my wall! Yeah, it's mostly for the personal challenge really.

It's "smart points", so you feel like you did well.

I know it totally doesn't matter, but I still feel kind of betrayed over the math test and only getting an 85.

I guess I have a different take on the grading -- I just don't care and am mostly in it for the learning with myself as the feedback for what has gone well. It probably has a lot to do with the fact that in school I always felt grades to be arbitrary and disconnected from how well I actually did (bad teachers and later professors probably having a lot to do with that). I'm mostly using Owltest for feedback on whether my methods have worked as expected. In the last assignment I think I got an 82 or something and I worked through the bugs up to that point to get it there. Once there I noticed what the problem was, realized it would just be busywork to fix it, and submitted because I knew fixing it wouldn't teach me anything new. That's actually why I like the Coursera model -- it gives me the freedom to do that while in school I'd have to copy whatever BS answer I knew the prof wanted me to regurgitate from the text.

Not saying this is better than using the grades themselves as feedback, it's just a way that works for me with the Coursera model and why I'm having fun with it so far. I love learning; I just hate school.

Also, I'd be annoyed too if that stupid math test was the combo breaker on my 100%'s.

So, uh, how long has it taken you guys to see a score reflected on the mini-project page?

Probably not as long as you resubmit?

Usually not more than a half-hour, Mantis.


Apparently it's a very strict hard deadline. First submission was 15:59:40, 15:59 is my local deadline.

Aww, that sucks LM.


Yeah, I guess I'm in this now for the edumucation and not the points. What a waste.

Stupid self-improvement ruins everything.

That really, really bites. I'm sorry, Mantis. I was worried about that all day, because I knew it was time-consuming, but I thought Coursera allowed a little slop in their deadlines.

Make sure to save your URLs, so you can resubmit your work when the course runs next time, if it does.

No worries. Yeah, I keep all my projects in a local git repo anyway.

Is anyone doing this through the capstone project? I think I'll just do this course for free which precludes the capstone.

Hmm, that idea based on getting to a cookie goal as fast as possible works, but it doesn't seem to work as well as buying for efficiency, which confuses me. That algorithm is only getting to 1.292e18, as opposed to the 1.314e18 that the 'efficient' strategy gets.

Maybe there's a bug in my implementation. I'll have to keep thinking about it.

I got my Wk1 homework finished up. I can't say I'm a big fan of these "enter the mathematical expression" questions. I fought question 9 for

def trial(n): val = random.randrange(n) return val

and still only received partial credit.

If it's not too late, do what I did when I first got partial credit: I tested it. I ran a million iterations, summed them, and divided by a million.

Even I was able to write the equation once I saw the actual value being generated.

Nah, I'm out of submissions now. It should effectively be (n-1) / 2, right? Or ( n(n-1)/2 ) / (n-1)

Oooooh, it should be ( n(n-1)/2 ) / n because the sum is 0+1...+(n-1) but it's still n possibilities.

[Edit] I just came in from baking my brain from a bike ride in 100º weather, I need to quit trying my rusty hand at math because that still simplifies down to (n-1)/2.

But it's a dry heat.

Yeah, (n-1) / 2 ... Coursera's quiz engine reduces that to something weird, but it comes back as the correct answer.

Oh, edit, I just saw this from awhile back:

I think I'll just do this course for free which precludes the capstone.

From what they were saying in the course notes, you'll be able to do the capstone even if you're a free student, if you want to.

I've been putting some time in the last couple of days into The Python Challenge. Fun riddles to solve with programming, beats taking yet another course! I'm only at exercise 5, but I've already had to learn new stuff and apply it all on my own, most interestingly regular expressions. Bit of a head-first dive, but certainly a great way to expand my knowledge and get familiar with the Python docs. Anyone here try this?

How are you guys iterating through a range? The stupid pylint is forcing me to use while loops and a manual index.


my_idx = 0 while my_idx < len(some_count): # do stuff my_idx += 1

rather than

for unused_i in range(some_count): # do stuff

Oh, I'm also curious how you guys initiate your scores list of lists, which is related to this question. I don't particularly love my version.

board_size = board.get_dim() scores_row = [] setup_idx = 0 while setup_idx < board_size: scores_row.append(0) setup_idx += 1 scores = board_size * [list(scores_row)]

[Edit] Ah-ha! Their pylint config supports dummy and _fooVar as unused variables, where fooVar can be anything prefixed with an underscore.

Man, my player is dumb and I don't know what's up. I get penalized for poor moves in the mc_move function. All my unit tests pass through all the test suites I found and my debug prints all look good, but for some reason he plays poorly.

[Edit] Well, I still don't see what was wrong with my prior version, but I refactored my get_best_move function to walk through the empty squares and compare each to max, updating as appropriate, and got it to work. My prior version used a dictionary with scores as keys with a list of the squares for that score. I grabbed the max score from the keys then randomly chose from the list. Unfortunately it chose poorly. The funny thing is that I wanted to just do what I ended up doing at the beginning but figured it was a cool use of dictionaries.