|
Thread Rules
1. This is not a "do my homework for me" thread. If you have specific questions, ask, but don't post an assignment or homework problem and expect an exact solution.
2. No recruiting for your cockamamie projects (you won't replace facebook with 3 dudes you found on the internet and $20)
3. If you can't articulate why a language is bad, don't start slinging shit about it. Just remember that nothing is worse than making CSS IE6 compatible.
4. Use [code] tags to format code blocks. |
I have an assignment for my AI class, part of the assignment is doing english to first order logic to CNF
Here are the english statements:
1.) For every positive integer, there is a larger positive integer.
2.) 1 is the smallest positive integer
3.) There are at least two distinct positive integers
And they provided established predicates:
PosInt(x) - indicates x is a positive integer
LessThan(x,y) - indicates than x is less than y
Equals(x,y) - indicates that the value of x equals the value of y
I would like to check if this is correct... at all...honestly the lecture and notes do a horrible job explaining how to do this
FOL:
1.) for all x, exists y(PosInt(x) --> PosInt(y)(LessThan(x,y))
2.) for all x(PosInt(x) ----> !LessThan(x,1))
3.) for all x, exists y(PosInt(x) ----> !Equals(x,y))
I guess I'll stop here for now, and worry about CNF if that is actually correct
|
Hey travis,
1) In your first example you are missing a closing bracket. You should also use an AND between "PosInt(y)" and "(LessThan(x,y)"
2) Technically, what you are saying is: "There is no positive integer that is smaller than 1". I assume you can assume it is proven that this means, that 1 is the smallest one, but I don't know how nitpicky your prof or tutor is. To be extra safe I would use a combination of NOT and EXISTS
3) Technically you are saying: "For every positive integer, there is at least one different positive integer". This is more than what was asked for. And again, you can assume that it is proven that one follows the other, mathematically. But if this is an excercise in using logic to model english sentences, than your answer is not correct. Two EXISTS should suffice here.
Also you wrote "Equals(x,y) - indicates that the value of x is less than the value of y". I hope this is a copy&paste error on your side. Otherwise, your prof or tutor is an ass. 
I hope this could help you. I didn't want to go too far not to break the "Do my homework" rule.
Cheers!
Edit: Also your notation was a little strange to me. Is this how they teach you notation?
|
Other than using english instead of qualifiers the notation seems to be the same as is posted in our notes.
For #1, I see what you are saying.
What about:
for all x, exists y(PosInt(x) ----> LessThan(x,PosInt(y)))
can I do that?
2.) yeah I saw that technicality. Hmm, not sure what to do then though
3.) are you saying to just do:
exists x, exists y(PosInt(x) ----> !Equals(x,y)) ?
If so, that makes sense to me
I wouldn't worry too much about the no homework rule, I'm not here to get free answers, just to understand things. It's tough when you live far from the school I think. I don't have any friends in my classes and it can be hard to go to office hours.
Also, fixed the error. Thank you!
EDIT: Actually, for #3, can I just do exists x, exists y, !Equals(PosInt(x),PosInt(y)) ?
|
Predicates arent nestable like that. What he meant was:
for all x, exists y(PosInt(x) ----> (PosInt(y) AND LessThan(x,y)))
Yes your edit for #3 works if you dont nest the predicates but use an AND
|
|
|
I'm making a scraper that tries to look at the big supermarkets in my country and download the images with the product name as the file name.
Initially it worked fine. I worked on my script and I did 2 supermarkets without too much problems. Then I tried another one and I couldn't access the html blocks in the main class, while other chunks were still accessible (on the same page)
Here's where I have to admit something: I don't know much about the rules and ethics. For instance, I just found out there's a robots.txt file for many (if not almost all) websites and see that the specific page and everything after it is blocked. However, I can and am still going to the page, but I can't see all the elements. Are those elements the only ones being given up 'for free' if I ignore the site's guidelines anyway. I don't really know what's going on. Is there a way to make me seem human and still get access? I don't want to slow anything down, I've learned you need to put pauses in your requests, but I feel like it would be an enormous waste of product info if I can't get those products from that specific site lol
By the way: this project is for personal use (would love to know all the products that are gluten free so my girlfriend who suffers from coeliac disease doesn't need to endlessly look at every product's ingredients etc).
|
Hyrule19194 Posts
If you are ignoring robots.txt you're a terrible person by definition.
Anyway, if stuff is being loaded dynamically (e.g. by js libraries loading stuff asynchronously) and your scraper doesn't do that, you won't get that information.
|
Germany2687 Posts
You could try a headless browser like puppeteer.
|
|
|
Thanks for the replies, I'm an absolute noob trying to do and learn along the way.
I'm using the requests and requests HTLM library for python.
I'll look into puppeteer, but that might be above me already lol
|
|
|
doing a project for a class
in the code the comments says this:
Parameters:
box_target : (tensor) size (-1, 5)
box_pred : (tensor) size (-1, 5)
What does this mean??
I've seen it before and it's generally just like a 5 index array I think. I just don't actually know wtf the -1 does to the shape.
|
On May 05 2019 08:27 travis wrote:
doing a project for a class
in the code the comments says this:
Parameters:
box_target : (tensor) size (-1, 5)
box_pred : (tensor) size (-1, 5)
What does this mean??
I've seen it before and it's generally just like a 5 index array I think. I just don't actually know wtf the -1 does to the shape.
Probably means it can be any size. What framework are you working in?
|
Zurich15363 Posts
|
Anyone here have experience using YOLO object detector?
It's my current class project and we are supposed to basically fill in the missing code, but I feel like the code makes no sense and the hints make no sense.
It's this if anyone understands and can take a look at it.
http://slazebni.cs.illinois.edu/fall18/assignment3_part2.html
(we are supposed to implement yolo_loss.py)
I am mostly confused at the part where we are supposed to compute the IOU
the shapes don't make any sense to me... and I can't tell what I am finding the "max IOU" of. It says "of the two boxes". But of what two boxes, the two boxes per grid cell? If I am doing that then why are the tensors shaped in a way that is completely counterproductive to that?
It doesn't help that all the "hints" are broken english. This is by far the most frustrating assignment I have been given. I kind of want to just ignore all the hints and directions and implement it the way I want.
|
Anyone here been to the VS Live events? I might be able to go this year, curious if they're worth the time.
|
I want to ask a few questions here about LSTMs. Is anyone here familiar with recurrent networks? The discussion may not need to be on LSTMs in particular.
|
What are you wondering about? I'm no expert on it, but I have been applying it quite a lot in custom named entity recognition and sentence classification in the last 8 month at work.
|
I have some confusion about how the recurrent part of the model actually transfers information from one state to the next
Say we have a LSTM layer of a network, that takes in an input of size X, outputs size Y
During training or prediction, what does the first iteration do? It takes in input X but what about the input from the LSTM itself? Is that initialized as all zeros?
Also, the recurrent part of the lstm is a single vector that keeps something akin to a memory input of previous outputs, is that right? But how does that work when the input to the LSTM is discrete values that correspond to labels (like, types of things). Are all of the last K steps are somehow stored in a single vector of a size that does not change? How does that information get combined into a single vector without completely screwing up what the inputs mean?
Or is it actually stored in a 3d tensor which starts as zeros and one of the dimensions corresponds to number of remembered iterations? Thanks
I've tried to watch some videos on LSTMs but they don't really go into that.
I ask because I am trying to understand how alphastar would be using an LSTM, because they must be using it for either the entire sequence of the game or for a very long sequence, because otherwise it seems useless. But... I don't really understand how.
|
I've got this MS SQL script that I can't figure out. See below (generalized):merge TableTarget TGT
using (
select * from TableSource
) as SRC
on TGT.ColumnA = SRC.ColumnA
when matched and exists (
select SRC.ColulmnB, SRC.ColumnC
except
select TGT.ColumnB, TGT.ColumnC
)
then update ...
when not matched by target then insert ...
I don't know what "when matched and exists" means, nor what the "except" statement means. I know what they mean generally speaking, and I know what a merge statement is, but in this context I can't figure out what they're trying to do. Any help would be appreciated.
|
|
|
|
|
|
EPT
