Hey guys, I'm using MIT Scheme and trying to write a procedure to find the average of all the numbers in a bunch of nested lists, for example:
(average-lists (list 1 2 (list 3 (list 4 5)) 6)))
Should return 3.5. I've played with the following code for days, and right now I've got it returning the sum, but not the average. Also, it is important that the values of the inner-most lists are calculated first, so no extracting all values and simply averaging them.
Here's what I have so far:
(define (average-lists data)
(if (null? data)
0.0
(if (list? (car data))
(+ (average-lists (car data)) (average-lists (cdr data)))
(+ (car data) (average-lists (cdr data))))))
I've tried this approach, as well as trying to use map to map a lambda function to it recursively, and a few others, but I just can't find one. I think I'm making thing harder than it should be.
I wrote the following in an effort to pursue some other paths as well, which you may find useful:
(define (list-num? x) ;Checks to see if list only contains numbers
(= (length (filter number? x)) (length x)))
(define (list-avg x) ;Returns the average of a list of numbers
(/ (accumulate + 0 x) (length x)))
Your help is really appreciated! This problem has been a nightmare for me. :)
Unless the parameters require otherwise, you'll want to define a helper procedure that can calculate both the sum and the count of how many items are in each list. Once you can average a single list, it's easy to adapt it to nested lists by checking to see if the car is a list.
This method will get you the average in one pass over the list, rather than the two or more passes that solutions that flatten the list or do the count and the sums in two separate passes. You would have to get the sum and counts separately from the sublists to get the overall average, though (re. zinglon's comment below).
Edit:
One way to get both the sum and the count back is to pass it back in a pair:
(define sum-and-count ; returns (sum . count)
(lambda (ls)
(if (null? ls)
(cons 0 0)
(let ((r (sum-and-count (cdr ls))))
(cons (+ (car ls) (car r))
(add1 (cdr r)))))))
That procedure gets the sum and number of elements of a list. Do what you did to your own average-lists to it to get it to examine deeply-nested lists. Then you can get the average by doing (/ (car result) (cdr result)).
Or, you can write separate deep-sum and deep-count procedures, and then do (/ (deep-sum ls) (deep-count ls)), but that requires two passes over the list.
(define (flatten mylist)
(cond ((null? mylist) '())
((list? (car mylist)) (append (flatten (car mylist)) (flatten (cdr mylist))))
(else (cons (car mylist) (flatten (cdr mylist))))))
(define (myavg mylist)
(let ((flatlist (flatten mylist)))
(/ (apply + flatlist) (length flatlist))))
The first function flattens the list. That is, it converts '(1 2 (3 (4 5)) 6) to '(1 2 3 4 5 6)
Then its just a matter of applying + to the flat list and doing the average.
Reference for the first function:
http://www.dreamincode.net/code/snippet3229.htm
Related
Ultimately, i shall be trying to reimplement sorting algorithms in scheme for linked lists. I have written a subprocedure that will help me along the way. The goal is to simply swap 2 elements, given as arguments "pair1 and pair2" and then return the list.
(define (cons-til lst until)
(cond
((or (null? lst) (eq? (car lst) until)) '())
(else (cons (car lst) (cons-til (cdr lst) until)))))
(define (swap lst pair1 pair2)
(cons (cons (append (cons-til lst (car pair1))
(car pair2)) (car pair1)) (cdr pair2)))
(define my-list '(1 2 3 4 5 6 7))
(swap my-list (cdr (cdr my-list)) (cdr (cdr (cdr my-list))))
When the code is executed, it returns:
(((1 2 . 4) . 3) 5 6 7)
How can i fix this in order to have a plain scheme list. The element seems to have swapped correctly.
Two suggestions:
Do you really want to write n cdr calls to index the nth element? I recommend strongly using integer indexes (if you need them, that is).
Referring to elements by index in a linked list (i. e. “random access”) is not very efficient most of the time, especially when done in loops. I strongly recommend using either vectors or a better suited algorithm that doesn't need random access, e. g. merge sort.
(define (swap2 lst pair1 pair2)
(append (append (append (cons-til lst (car pair1))
(list (car pair2)))
(list (car pair1))) (cdr pair2)))
This code seems to work. I'm not sure this is completely efficient or a smart solution to the problem. Looking forward to other suggestions. The value given back is '(1 2 4 3 5 6 7)
I'm stuck on a homework question and could use any hints or suggestions. I need to find the n largest numbers in a list using Scheme. I am trying to do this by creating helper functions that are called by the main function. So far I have this:
(define (get_max_value L)
(if (null? L)
'()
(apply max L)
)
(define (biggest_nums L n)
(if (null? n)
'()
(cons (get_max_value L) (biggest_nums L (- n 1)))
)
)
When I type (biggest_num '(3 1 4 2 5) 3) at the command prompt drRacket just hangs and doesn't even return an error message. Where am I going wrong?
The simplest solution is to first sort the numbers in ascending order and then take the n first. This translates quite literally in Racket code:
(define (biggest_nums L n)
(take (sort L >) n))
It works as expected:
(biggest_nums '(3 1 4 2 5) 3)
=> '(5 4 3)
Two mains problems with your code:
L always stays the same. L doesn't decrease in size when you make the recursive call, so the max will always be the same number in every recursive call.
You don't ever check n to make sure it contains the correct amount of numbers before returning the answer.
To solve these two problems in the most trivial way possible, you can put a (< n 1) condition in the if, and use something like (cdr L) to make L decrease in size in each recursive call by removing an element each time.
(define (biggest-nums n L)
(if (or (empty? L)
(< n 1))
'()
(cons (apply max L) (biggest-nums (- n 1) (cdr L)))))
So when we run it:
> (biggest-nums 3 '(1 59 2 10 33 4 5))
What should the output be?
'(59 33 10)
What is the actual output?
'(59 59 33)
OK, so we got your code running, but there are still some issues with it. Do you know why that's happening? Can you step through the code to figure out what you could do to fix it?
Just sort the list and then return the first n elements.
However, if the list is very long and n is not very large, then you probably don't want to sort the whole list first. In that case, I would suggest something like this:
(define insert-sorted
(lambda (item lst)
(cond ((null? lst)
(list item))
((<= item (car lst))
(cons item lst))
(else
(cons (car lst) (insert-sorted item (cdr lst)))))))
(define largest-n
(lambda (count lst)
(if (<= (length lst) count)
lst
(let loop ((todo (cdr lst))
(result (list (car lst))))
(if (null? todo)
result
(let* ((item (car todo))
(new-result
(if (< (car result) item)
(let ((new-result (insert-sorted item result)))
(if (< count (length new-result))
(cdr new-result)
new-result))
result)))
(loop (cdr todo)
new-result)))))))
Pretty straightforward question. My initial approach was to define another procedure to find the last element of lst within first-last. After finding the last element I appended it with the first element of lst (car lst).
This is how append works.
(append list1 list2)
e.g., (append '(1 2 3) '(2 1 5)) -> (1 2 3 2 1 5)
I'm wondering if the problem is just with my syntax but I am not sure.
(define (first-last lst)
(define (last lst)
(cond ((null? (cdr lst))(car lst))
(else (last (cdr lst)))))
(append(car lst)(last lst)))
The error occurs in the
(append(car lst)(last lst)))
"mcar: contract violation
expected: mpair?
given: 1"
This is my first question on stack, so I'm sorry if the question is not presented in the correct way.
append is only for joining two or more lists. Here, though, you're not joining existing lists, but building a list from two elements. For that, use list:
(list (car lst) (last lst))
If you can use match, a neat solution is possible:
(define first-last
(lambda (x)
(match x
((first rest ... last)
(list first last))
((only) (list only only))
(_ #f))))
Of course, you could return something other than #f in the catch-all clause.
I've been trying to solve exercise 2.20 of SICP, where "dotted-tail" notation is introduced. My problem is that, instead of returning a proper list with results, my function returns a nested list.
I know that there is something wrong with the way I'm calling cons, but I still have no clue of how to solve the issue.
So here is my function:
(define (same-parity first . items)
(define (speccar items)
(cond ((null? items) 2)
((not (pair? items)) (modulo items 2))
(else (modulo (car items) 2))))
(define (iter-parity first items result)
(let ((parityFirst (modulo first 2)) (samepar (speccar items)))
(if (null? items)
result
(if (= parityFirst samepar)
;; This next line is where the problem is...
(iter-parity first (cdr items) (cons (list result) (list (car items))))
(iter-parity first (cdr items) result)))))
(iter-parity first items first))
Test:
(same-parity 1 2 3 4 5)
((((1) 3)) 5)
Now, I've read the following answers that deal with a similar problem:
Cons element to list vs cons list to element in Scheme
How to use 'cons' without generating nested lists in Scheme?
They certainly make it clear where the problem is coming from, but how does one go about to actually implement a proper solution?
And, if possible, what is the correct way of "thinking" in Scheme to avoid these traps/pitfalls?
You're incorrectly building the output list - remember: the first argument to cons should be the current element and the second argument, the result list.
Also, given that you're using tail recursion, you'll have to reverse the output at the end to preserve the same order as in the original list. Try this:
(define (same-parity first . items)
(define (speccar items)
(cond ((null? items) 2)
((not (pair? items)) (modulo items 2))
(else (modulo (car items) 2))))
(define (iter-parity first items result)
(let ((parityFirst (modulo first 2))
(samepar (speccar items)))
(if (null? items)
(reverse result)
(if (= parityFirst samepar)
(iter-parity first
(cdr items)
(cons (car items) result))
(iter-parity first
(cdr items)
result)))))
(iter-parity first items (list first)))
The above solution can be greatly simplified if we use built-in procedures (don't reinvent the wheel!). This is the recommended way to write programs in Scheme - adhering to a functional style:
(define (same-parity head . tail)
(if (even? head)
(filter even? (cons head tail))
(filter odd? (cons head tail))))
Either way, it works as expected:
(same-parity 1 2 3 4 5)
=> '(1 3 5)
Write a procedure (first-half lst) that returns a list with the first half of its elements. If the length of the given list is odd, the returned list should have (length - 1) / 2 elements.
I am given these program as a example and as I am new to Scheme I need your help in solving this problem.
(define list-head
(lambda (lst k)
(if (= k 0)
'()
(cons (car lst)(list-head (cdr lst)(- k 1)))))))
(list-head '(0 1 2 3 4) 3)
; list the first 3 element in the list (list 0 1 2)
Also the expected output for the program I want is :
(first-half '(43 23 14 5 9 57 0 125))
(43 23 14 5)
This is pretty simple to implement in terms of existing procedures, check your interpreter's documentation for the availability of the take procedure:
(define (first-half lst)
(take lst (quotient (length lst) 2)))
Apart from that, the code provided in the question is basically reinventing take, and it looks correct. The only detail left to implement would be, how to obtain the half of the lists' length? same as above, just use the quotient procedure:
(define (first-half lst)
(list-head lst (quotient (length lst) 2)))
It looks like you are learning about recursion? One recursive approach is to walk the list with a 'slow' and 'fast' pointer; when the fast pointer reaches the end you are done; use the slow pointer to grow the result. Like this:
(define (half list)
(let halving ((rslt '()) (slow list) (fast list))
(if (or (null? fast) (null? (cdr fast)))
(reverse rslt)
(halving (cons (car slow) rslt)
(cdr slow)
(cdr (cdr fast))))))
Another way to approach it is to have a function that divides the list at a specific index, and then a wrapper to calculate floor(length/2):
(define (cleave_at n a)
(cond
((null? a) '())
((zero? n) (list '() a))
(#t
((lambda (x)
(cons (cons (car a) (car x)) (cdr x)))
(cleave_at (- n 1) (cdr a))))))
(define (first-half a)
(car (cleave_at (floor (/ (length a) 2)) a)))