Related
Q. Given two arrays, A and B, of equal length, find the largest possible contiguous subarray of indices [i,j] such that max(A[i: j]) < min(B[i: j]).
Example: A = [10, 21, 5, 1, 3], B = [3, 1, 4, 23, 56]
Explanation: A[4] = 1, A[5] = 3, B[4] = 23, B[5] = 56, max(A[4], A[5]) < min(B[4], B[5])
The indices are [4,5] (inclusive), and the largest contiguous subarray has length 2
I can do this in O(n2) brute force method but cannot seem to reduce the time complexity. Any ideas?
O(n) solution:
Move index j from left to right and drag i behind so that the window from i to j is valid. So always increase j by 1, and then increase i as much as needed for the window to be valid.
To do that, keep a queue Aq of indexes of non-increasing A-values. Then A[Aq[0]] tells you the max A-value in the window. Similarly, keep a queue for non-decreasing B-values.
For each new j, first update Aq and Bq according to the new A-value and new B-value. Then, while the window is invalid, increase i and drop Aq[0] and Bq[0] if they're i. When both j and i are updated, update the result with the window size j - i - 1.
Python implementation:
def solution(A, B):
Aq = deque()
Bq = deque()
i = 0
maxlen = 0
for j in range(len(A)):
while Aq and A[Aq[-1]] < A[j]:
Aq.pop()
Aq.append(j)
while Bq and B[Bq[-1]] > B[j]:
Bq.pop()
Bq.append(j)
while Aq and A[Aq[0]] >= B[Bq[0]]:
if Aq[0] == i:
Aq.popleft()
if Bq[0] == i:
Bq.popleft()
i += 1
maxlen = max(maxlen, j - i + 1)
return maxlen
Test results from comparing against a naive brute force reference solution:
expect: 83 result: 83 same: True
expect: 147 result: 147 same: True
expect: 105 result: 105 same: True
expect: 71 result: 71 same: True
expect: 110 result: 110 same: True
expect: 56 result: 56 same: True
expect: 140 result: 140 same: True
expect: 109 result: 109 same: True
expect: 86 result: 86 same: True
expect: 166 result: 166 same: True
Testing code (Try it online!)
from timeit import timeit
from random import choices
from collections import deque
from itertools import combinations
def solution(A, B):
Aq = deque()
Bq = deque()
i = 0
maxlen = 0
for j in range(len(A)):
while Aq and A[Aq[-1]] < A[j]:
Aq.pop()
Aq.append(j)
while Bq and B[Bq[-1]] > B[j]:
Bq.pop()
Bq.append(j)
while Aq and A[Aq[0]] >= B[Bq[0]]:
if Aq[0] == i:
Aq.popleft()
if Bq[0] == i:
Bq.popleft()
i += 1
maxlen = max(maxlen, j - i + 1)
return maxlen
def naive(A, B):
return max((j - i + 1
for i, j in combinations(range(len(A)), 2)
if max(A[i:j+1]) < min(B[i:j+1])),
default=0)
for _ in range(10):
n = 500
A = choices(range(42), k=n)
B = choices(range(1234), k=n)
expect = naive(A, B)
result = solution(A, B)
print(f'expect: {expect:3} result: {result:3} same: {result == expect}')
I can see based on the problem, saying we have 2 conditions:
max(A[i,j-1]) < min(B[i,j-1])
max(A[i,j]) >= min(B[i,j])
saying maxA is index of max item in A array in [i,j], minB is index of min item in B array in [i,j]; and anchor is min(maxA, minB)
Then we will have: max(A[i+k,anchor]) >= min(B[i+k,anchor]) ∀ k in [i+1,anchor].
So I come up with a simple algorithm like below:
int extractLongestRange(int n, int[] A, int[] B) {
// n is size of both array
int size = 0;
for(int i = 0; i < n; i++){
int maxAIndex = i;
int minBIndex = i;
for(int j = i; j < n; j++){
if(A[maxAIndex] < A[j]){
maxAIndex = j;
}
if(B[minBIndex] > B[j]){
minBIndex = j;
}
if(A[maxAIndex] >= B[minBIndex]){
if(size < j - i){
size = j - i;
}
// here, need to jump back to min of maxAIndex and minBIndex.
i = Math.min(maxAIndex, minBIndex);
break;
}
// this case, if j reach the end of array
if(j == n - 1){
if(size < j - i + 1){
size = j - i + 1;
i = j;
}
}
}
}
return size;
}
With this approach, the complexity is O(n).
We can change the output to pick-up the other information if needed.
This can be solved in O(n log(n)).
Here is an outline of my technique.
My idea looks like a "rising water level" of maximum in A, keeping track of the "islands" emerging in A, and the "islands" submerging in B. And recording the maximum intersections after they appear from A emerging, or before they disappear from B sinking.
We will need 2 balanced binary trees of intervals in A and B, and a priority queue of events.
The tree of intervals need to support a logarithmic "find and return interval containing i if it exists". It also needs to support a logarithmic "add i to tree, extending/merging intervals as appropriate, and return new interval". And likewise a logarithmic "remove i from tree, shortening, splitting, or deleting its interval as appropriate".
Events can be of the form "remove B[i]" or "add A[i]". The priority queue is sorted first by the size of the element added/subtracted, and then by putting B before A. (So we do not elements of size x to A until all of the elements of size x are removed from B.)
With that in mind, here is pseudocode for how to do it.
Put all possible events in the priority queue
Initialize an empty tree of intervals A_tree
Initialize a tree of intervals B_tree containing the entire range
Initialize max interval to (0, -1) (length 0)
For each event (type, i) in queue:
if event.type = A:
new_interval = A_tree.add(event.i)
if event.i in B_tree:
Intersect event.i's A_tree interval with event.i's B_tree interval
if intersection is longer than max_interval:
update to new and better max_interval
else:
if event.i in A_tree:
Intersect event.i's A_tree interval with event.i's B_tree interval
if intersection is longer than max_interval:
update to new and better max_interval
B_tree.remove(event.i)
Processing any event is O(log(n). There are 2n = O(n) events. So overall time is O(n log(n)).
I have to find height of tree and find protection number (or just to generate a tree) from balanced parentheses.
For example:
()()()() creates tree like a list with height 3.
I have no idea how to convert parentheses to tree. I found some 'answers':
http://www.cs.utsa.edu/~wagner/knuth/fasc4a.pdf (second page contains all examples for tree with 4 nodes)
paragraph - Binary Trees, Forests, Non-Crossing Pairs :
https://sahandsaba.com/interview-question-generating-all-balanced-parentheses.html
However, I still don't know how to create a tree from such defined parentheses. I have some impression that in Knuth, authors treat it as something obvious.
Do I miss something or it's not that simple?
Is it necessary to create a forest and then a binary tree?
A pair of parentheses represents a node. What appears within those parentheses represents its left child's subtree (according to the same rules). What appears at the right of those parentheses represents the node's right child's subtree (again, according to the same rules).
The conversion of this encoding into a binary tree can be done recursively like this:
function makeBinaryTree(input):
i = 0 # character index in input
function recur():
if i >= input.length or input[i] == ")":
i = i + 1
return NIL
i = i + 1
node = new Node
node.left = recur()
if i >= input.length or input[i] == ")":
i = i + 1
return node
node.right = recur()
return node
return recur()
Here is an implementation in JavaScript that performs the conversion for each of those 4-noded trees, and pretty prints the resulting trees:
function makeBinaryTree(input) {
let i = 0; // character index in input
return recur();
function recur() {
if (i >= input.length || input[i++] === ")") return null;
let node = { left: recur(), right: null };
if (i >= input.length || input[i] === ")") {
i++;
return node;
}
node.right = recur();
return node;
}
}
// Helper function to pretty print a tree
const disc = "\u2B24";
function treeAsLines(node) {
let left = [""], right = [""];
if (node.left) left = treeAsLines(node.left);
if (node.right) right = treeAsLines(node.right);
while (left.length < right.length) left.push(" ".repeat(left[0].length));
while (left.length > right.length) right.push(" ".repeat(left[0].length));
let topLeft = "", topRight = "";
let i = left[0].indexOf(disc);
if (i > -1) topLeft = "┌".padEnd(left[0].length-i+1, "─");
i = right[0].indexOf(disc);
if (i > -1) topRight = "┐".padStart(i+2, "─");
return [topLeft.padStart(left[0].length+1) + disc + topRight.padEnd(right[0].length+1)]
.concat(left.map((line, i) => line + " " + right[i]));
}
// The trees as listed in Table 1 of http://www.cs.utsa.edu/~wagner/knuth/fasc4a.pdf
let inputs = [
"()()()()",
"()()(())",
"()(())()",
"()(()())",
"()((()))",
"(())()()",
"(())(())",
"(()())()",
"(()()())",
"(()(()))",
"((()))()",
"((())())",
"((()()))",
"(((())))"
];
for (let input of inputs) {
let tree = makeBinaryTree(input);
console.log(input);
console.log(treeAsLines(tree).join("\n"));
}
If I understood Knuth correctly, the representation works as the following: A pair of matching parentheses represents a node, e.g. () = A. Two consecutive pairs of matching parentheses means that the second node is the right child of the first one, e.g. ()() = A -> B. And two pairs of embedded parentheses means the inside node is the left child of the outside node, i.e. (()) = B <- A. Therefore, ()()()() = A -> B -> C -> D.
A possible algorithm to convert parentheses to binary tree would be:
convert(parentheses):
if parentheses is empty:
return Nil
root = Node()
left_start = 1
left_end = Nil
open = 0
for p = 0 to |parentheses|-1:
if parentheses[p] == '(':
open += 1
else
open -= 1
if open == 0:
left_end = p
break
root.left = convert(parentheses[left_start:left_end] or empty if index out of bound)
root.right = convert(parentheses[left_end+1:] or empty if index out of bound)
return root
It works by converting parentheses (L)R in the binary tree L <- A -> R recursively.
I'm trying to solve DNA problem which is more of improved(?) version of LCS problem.
In the problem, there is string which is string and semi-substring which allows part of string to have one or no letter skipped. For example, for string "desktop", it has semi-substring {"destop", "dek", "stop", "skop","desk","top"}, all of which has one or no letter skipped.
Now, I am given two DNA strings consisting of {a,t,g,c}. I"m trying to find longest semi-substring, LSS. and if there is more than one LSS, print out the one in the fastest order.
For example, two dnas {attgcgtagcaatg, tctcaggtcgatagtgac} prints out "tctagcaatg"
and aaaattttcccc, cccgggggaatatca prints out "aattc"
I'm trying to use common LCS algorithm but cannot solve it with tables although I did solve the one with no letter skipped. Any advice?
This is a variation on the dynamic programming solution for LCS, written in Python.
First I'm building up a Suffix Tree for all the substrings that can be made from each string with the skip rule. Then I'm intersecting the suffix trees. Then I'm looking for the longest string that can be made from that intersection tree.
Please note that this is technically O(n^2). Its worst case is when both strings are the same character, repeated over and over again. Because you wind up with a lot of what logically is something like, "an 'l' at position 42 in the one string could have matched against position l at position 54 in the other". But in practice it will be O(n).
def find_subtree (text, max_skip=1):
tree = {}
tree_at_position = {}
def subtree_from_position (position):
if position not in tree_at_position:
this_tree = {}
if position < len(text):
char = text[position]
# Make sure that we've populated the further tree.
subtree_from_position(position + 1)
# If this char appeared later, include those possible matches.
if char in tree:
for char2, subtree in tree[char].iteritems():
this_tree[char2] = subtree
# And now update the new choices.
for skip in range(max_skip + 1, 0, -1):
if position + skip < len(text):
this_tree[text[position + skip]] = subtree_from_position(position + skip)
tree[char] = this_tree
tree_at_position[position] = this_tree
return tree_at_position[position]
subtree_from_position(0)
return tree
def find_longest_common_semistring (text1, text2):
tree1 = find_subtree(text1)
tree2 = find_subtree(text2)
answered = {}
def find_intersection (subtree1, subtree2):
unique = (id(subtree1), id(subtree2))
if unique not in answered:
answer = {}
for k, v in subtree1.iteritems():
if k in subtree2:
answer[k] = find_intersection(v, subtree2[k])
answered[unique] = answer
return answered[unique]
found_longest = {}
def find_longest (tree):
if id(tree) not in found_longest:
best_candidate = ''
for char, subtree in tree.iteritems():
candidate = char + find_longest(subtree)
if len(best_candidate) < len(candidate):
best_candidate = candidate
found_longest[id(tree)] = best_candidate
return found_longest[id(tree)]
intersection_tree = find_intersection(tree1, tree2)
return find_longest(intersection_tree)
print(find_longest_common_semistring("attgcgtagcaatg", "tctcaggtcgatagtgac"))
Let g(c, rs, rt) represent the longest common semi-substring of strings, S and T, ending at rs and rt, where rs and rt are the ranked occurences of the character, c, in S and T, respectively, and K is the number of skips allowed. Then we can form a recursion which we would be obliged to perform on all pairs of c in S and T.
JavaScript code:
function f(S, T, K){
// mapS maps a char to indexes of its occurrences in S
// rsS maps the index in S to that char's rank (index) in mapS
const [mapS, rsS] = mapString(S)
const [mapT, rsT] = mapString(T)
// h is used to memoize g
const h = {}
function g(c, rs, rt){
if (rs < 0 || rt < 0)
return 0
if (h.hasOwnProperty([c, rs, rt]))
return h[[c, rs, rt]]
// (We are guaranteed to be on
// a match in this state.)
let best = [1, c]
let idxS = mapS[c][rs]
let idxT = mapT[c][rt]
if (idxS == 0 || idxT == 0)
return best
for (let i=idxS-1; i>=Math.max(0, idxS - 1 - K); i--){
for (let j=idxT-1; j>=Math.max(0, idxT - 1 - K); j--){
if (S[i] == T[j]){
const [len, str] = g(S[i], rsS[i], rsT[j])
if (len + 1 >= best[0])
best = [len + 1, str + c]
}
}
}
return h[[c, rs, rt]] = best
}
let best = [0, '']
for (let c of Object.keys(mapS)){
for (let i=0; i<(mapS[c]||[]).length; i++){
for (let j=0; j<(mapT[c]||[]).length; j++){
let [len, str] = g(c, i, j)
if (len > best[0])
best = [len, str]
}
}
}
return best
}
function mapString(s){
let map = {}
let rs = []
for (let i=0; i<s.length; i++){
if (!map[s[i]]){
map[s[i]] = [i]
rs.push(0)
} else {
map[s[i]].push(i)
rs.push(map[s[i]].length - 1)
}
}
return [map, rs]
}
console.log(f('attgcgtagcaatg', 'tctcaggtcgatagtgac', 1))
console.log(f('aaaattttcccc', 'cccgggggaatatca', 1))
console.log(f('abcade', 'axe', 1))
I have always wanted to do this but every time I start thinking about the problem it blows my mind because of its exponential nature.
The problem solver I want to be able to understand and code is for the countdown maths problem:
Given set of number X1 to X5 calculate how they can be combined using mathematical operations to make Y.
You can apply multiplication, division, addition and subtraction.
So how does 1,3,7,6,8,3 make 348?
Answer: (((8 * 7) + 3) -1) *6 = 348.
How to write an algorithm that can solve this problem? Where do you begin when trying to solve a problem like this? What important considerations do you have to think about when designing such an algorithm?
Very quick and dirty solution in Java:
public class JavaApplication1
{
public static void main(String[] args)
{
List<Integer> list = Arrays.asList(1, 3, 7, 6, 8, 3);
for (Integer integer : list) {
List<Integer> runList = new ArrayList<>(list);
runList.remove(integer);
Result result = getOperations(runList, integer, 348);
if (result.success) {
System.out.println(integer + result.output);
return;
}
}
}
public static class Result
{
public String output;
public boolean success;
}
public static Result getOperations(List<Integer> numbers, int midNumber, int target)
{
Result midResult = new Result();
if (midNumber == target) {
midResult.success = true;
midResult.output = "";
return midResult;
}
for (Integer number : numbers) {
List<Integer> newList = new ArrayList<Integer>(numbers);
newList.remove(number);
if (newList.isEmpty()) {
if (midNumber - number == target) {
midResult.success = true;
midResult.output = "-" + number;
return midResult;
}
if (midNumber + number == target) {
midResult.success = true;
midResult.output = "+" + number;
return midResult;
}
if (midNumber * number == target) {
midResult.success = true;
midResult.output = "*" + number;
return midResult;
}
if (midNumber / number == target) {
midResult.success = true;
midResult.output = "/" + number;
return midResult;
}
midResult.success = false;
midResult.output = "f" + number;
return midResult;
} else {
midResult = getOperations(newList, midNumber - number, target);
if (midResult.success) {
midResult.output = "-" + number + midResult.output;
return midResult;
}
midResult = getOperations(newList, midNumber + number, target);
if (midResult.success) {
midResult.output = "+" + number + midResult.output;
return midResult;
}
midResult = getOperations(newList, midNumber * number, target);
if (midResult.success) {
midResult.output = "*" + number + midResult.output;
return midResult;
}
midResult = getOperations(newList, midNumber / number, target);
if (midResult.success) {
midResult.output = "/" + number + midResult.output;
return midResult
}
}
}
return midResult;
}
}
UPDATE
It's basically just simple brute force algorithm with exponential complexity.
However you can gain some improvemens by leveraging some heuristic function which will help you to order sequence of numbers or(and) operations you will process in each level of getOperatiosn() function recursion.
Example of such heuristic function is for example difference between mid result and total target result.
This way however only best-case and average-case complexities get improved. Worst case complexity remains untouched.
Worst case complexity can be improved by some kind of branch cutting. I'm not sure if it's possible in this case.
Sure it's exponential but it's tiny so a good (enough) naive implementation would be a good start. I suggest you drop the usual infix notation with bracketing, and use postfix, it's easier to program. You can always prettify the outputs as a separate stage.
Start by listing and evaluating all the (valid) sequences of numbers and operators. For example (in postfix):
1 3 7 6 8 3 + + + + + -> 28
1 3 7 6 8 3 + + + + - -> 26
My Java is laughable, I don't come here to be laughed at so I'll leave coding this up to you.
To all the smart people reading this: yes, I know that for even a small problem like this there are smarter approaches which are likely to be faster, I'm just pointing OP towards an initial working solution. Someone else can write the answer with the smarter solution(s).
So, to answer your questions:
I begin with an algorithm that I think will lead me quickly to a working solution. In this case the obvious (to me) choice is exhaustive enumeration and testing of all possible calculations.
If the obvious algorithm looks unappealing for performance reasons I'll start thinking more deeply about it, recalling other algorithms that I know about which are likely to deliver better performance. I may start coding one of those first instead.
If I stick with the exhaustive algorithm and find that the run-time is, in practice, too long, then I might go back to the previous step and code again. But it has to be worth my while, there's a cost/benefit assessment to be made -- as long as my code can outperform Rachel Riley I'd be satisfied.
Important considerations include my time vs computer time, mine costs a helluva lot more.
A working solution in c++11 below.
The basic idea is to use a stack-based evaluation (see RPN) and convert the viable solutions to infix notation for display purposes only.
If we have N input digits, we'll use (N-1) operators, as each operator is binary.
First we create valid permutations of operands and operators (the selector_ array). A valid permutation is one that can be evaluated without stack underflow and which ends with exactly one value (the result) on the stack. Thus 1 1 + is valid, but 1 + 1 is not.
We test each such operand-operator permutation with every permutation of operands (the values_ array) and every combination of operators (the ops_ array). Matching results are pretty-printed.
Arguments are taken from command line as [-s] <target> <digit>[ <digit>...]. The -s switch prevents exhaustive search, only the first matching result is printed.
(use ./mathpuzzle 348 1 3 7 6 8 3 to get the answer for the original question)
This solution doesn't allow concatenating the input digits to form numbers. That could be added as an additional outer loop.
The working code can be downloaded from here. (Note: I updated that code with support for concatenating input digits to form a solution)
See code comments for additional explanation.
#include <iostream>
#include <vector>
#include <algorithm>
#include <stack>
#include <iterator>
#include <string>
namespace {
enum class Op {
Add,
Sub,
Mul,
Div,
};
const std::size_t NumOps = static_cast<std::size_t>(Op::Div) + 1;
const Op FirstOp = Op::Add;
using Number = int;
class Evaluator {
std::vector<Number> values_; // stores our digits/number we can use
std::vector<Op> ops_; // stores the operators
std::vector<char> selector_; // used to select digit (0) or operator (1) when evaluating. should be std::vector<bool>, but that's broken
template <typename T>
using Stack = std::stack<T, std::vector<T>>;
// checks if a given number/operator order can be evaluated or not
bool isSelectorValid() const {
int numValues = 0;
for (auto s : selector_) {
if (s) {
if (--numValues <= 0) {
return false;
}
}
else {
++numValues;
}
}
return (numValues == 1);
}
// evaluates the current values_ and ops_ based on selector_
Number eval(Stack<Number> &stack) const {
auto vi = values_.cbegin();
auto oi = ops_.cbegin();
for (auto s : selector_) {
if (!s) {
stack.push(*(vi++));
continue;
}
Number top = stack.top();
stack.pop();
switch (*(oi++)) {
case Op::Add:
stack.top() += top;
break;
case Op::Sub:
stack.top() -= top;
break;
case Op::Mul:
stack.top() *= top;
break;
case Op::Div:
if (top == 0) {
return std::numeric_limits<Number>::max();
}
Number res = stack.top() / top;
if (res * top != stack.top()) {
return std::numeric_limits<Number>::max();
}
stack.top() = res;
break;
}
}
Number res = stack.top();
stack.pop();
return res;
}
bool nextValuesPermutation() {
return std::next_permutation(values_.begin(), values_.end());
}
bool nextOps() {
for (auto i = ops_.rbegin(), end = ops_.rend(); i != end; ++i) {
std::size_t next = static_cast<std::size_t>(*i) + 1;
if (next < NumOps) {
*i = static_cast<Op>(next);
return true;
}
*i = FirstOp;
}
return false;
}
bool nextSelectorPermutation() {
// the start permutation is always valid
do {
if (!std::next_permutation(selector_.begin(), selector_.end())) {
return false;
}
} while (!isSelectorValid());
return true;
}
static std::string buildExpr(const std::string& left, char op, const std::string &right) {
return std::string("(") + left + ' ' + op + ' ' + right + ')';
}
std::string toString() const {
Stack<std::string> stack;
auto vi = values_.cbegin();
auto oi = ops_.cbegin();
for (auto s : selector_) {
if (!s) {
stack.push(std::to_string(*(vi++)));
continue;
}
std::string top = stack.top();
stack.pop();
switch (*(oi++)) {
case Op::Add:
stack.top() = buildExpr(stack.top(), '+', top);
break;
case Op::Sub:
stack.top() = buildExpr(stack.top(), '-', top);
break;
case Op::Mul:
stack.top() = buildExpr(stack.top(), '*', top);
break;
case Op::Div:
stack.top() = buildExpr(stack.top(), '/', top);
break;
}
}
return stack.top();
}
public:
Evaluator(const std::vector<Number>& values) :
values_(values),
ops_(values.size() - 1, FirstOp),
selector_(2 * values.size() - 1, 0) {
std::fill(selector_.begin() + values_.size(), selector_.end(), 1);
std::sort(values_.begin(), values_.end());
}
// check for solutions
// 1) we create valid permutations of our selector_ array (eg: "1 1 + 1 +",
// "1 1 1 + +", but skip "1 + 1 1 +" as that cannot be evaluated
// 2) for each evaluation order, we permutate our values
// 3) for each value permutation we check with each combination of
// operators
//
// In the first version I used a local stack in eval() (see toString()) but
// it turned out to be a performance bottleneck, so now I use a cached
// stack. Reusing the stack gives an order of magnitude speed-up (from
// 4.3sec to 0.7sec) due to avoiding repeated allocations. Using
// std::vector as a backing store also gives a slight performance boost
// over the default std::deque.
std::size_t check(Number target, bool singleResult = false) {
Stack<Number> stack;
std::size_t res = 0;
do {
do {
do {
Number value = eval(stack);
if (value == target) {
++res;
std::cout << target << " = " << toString() << "\n";
if (singleResult) {
return res;
}
}
} while (nextOps());
} while (nextValuesPermutation());
} while (nextSelectorPermutation());
return res;
}
};
} // namespace
int main(int argc, const char **argv) {
int i = 1;
bool singleResult = false;
if (argc > 1 && std::string("-s") == argv[1]) {
singleResult = true;
++i;
}
if (argc < i + 2) {
std::cerr << argv[0] << " [-s] <target> <digit>[ <digit>]...\n";
std::exit(1);
}
Number target = std::stoi(argv[i]);
std::vector<Number> values;
while (++i < argc) {
values.push_back(std::stoi(argv[i]));
}
Evaluator evaluator{values};
std::size_t res = evaluator.check(target, singleResult);
if (!singleResult) {
std::cout << "Number of solutions: " << res << "\n";
}
return 0;
}
Input is obviously a set of digits and operators: D={1,3,3,6,7,8,3} and Op={+,-,*,/}. The most straight forward algorithm would be a brute force solver, which enumerates all possible combinations of these sets. Where the elements of set Op can be used as often as wanted, but elements from set D are used exactly once. Pseudo code:
D={1,3,3,6,7,8,3}
Op={+,-,*,/}
Solution=348
for each permutation D_ of D:
for each binary tree T with D_ as its leafs:
for each sequence of operators Op_ from Op with length |D_|-1:
label each inner tree node with operators from Op_
result = compute T using infix traversal
if result==Solution
return T
return nil
Other than that: read jedrus07's and HPM's answers.
By far the easiest approach is to intelligently brute force it. There is only a finite amount of expressions you can build out of 6 numbers and 4 operators, simply go through all of them.
How many? Since you don't have to use all numbers and may use the same operator multiple times, This problem is equivalent to "how many labeled strictly binary trees (aka full binary trees) can you make with at most 6 leaves, and four possible labels for each non-leaf node?".
The amount of full binary trees with n leaves is equal to catalan(n-1). You can see this as follows:
Every full binary tree with n leaves has n-1 internal nodes and corresponds to a non-full binary tree with n-1 nodes in a unique way (just delete all the leaves from the full one to get it). There happen to be catalan(n) possible binary trees with n nodes, so we can say that a strictly binary tree with n leaves has catalan(n-1) possible different structures.
There are 4 possible operators for each non-leaf node: 4^(n-1) possibilities
The leaves can be numbered in n! * (6 choose (n-1)) different ways. (Divide this by k! for each number that occurs k times, or just make sure all numbers are different)
So for 6 different numbers and 4 possible operators you get Sum(n=1...6) [ Catalan(n-1) * 6!/(6-n)! * 4^(n-1) ] possible expressions for a total of 33,665,406. Not a lot.
How do you enumerate these trees?
Given a collection of all trees with n-1 or less nodes, you can create all trees with n nodes by systematically pairing all of the n-1 trees with the empty tree, all n-2 trees with the 1 node tree, all n-3 trees with all 2 node tree etc. and using them as the left and right sub trees of a newly formed tree.
So starting with an empty set you first generate the tree that has just a root node, then from a new root you can use that either as a left or right sub tree which yields the two trees that look like this: / and . And so on.
You can turn them into a set of expressions on the fly (just loop over the operators and numbers) and evaluate them as you go until one yields the target number.
I've written my own countdown solver, in Python.
Here's the code; it is also available on GitHub:
#!/usr/bin/env python3
import sys
from itertools import combinations, product, zip_longest
from functools import lru_cache
assert sys.version_info >= (3, 6)
class Solutions:
def __init__(self, numbers):
self.all_numbers = numbers
self.size = len(numbers)
self.all_groups = self.unique_groups()
def unique_groups(self):
all_groups = {}
all_numbers, size = self.all_numbers, self.size
for m in range(1, size+1):
for numbers in combinations(all_numbers, m):
if numbers in all_groups:
continue
all_groups[numbers] = Group(numbers, all_groups)
return all_groups
def walk(self):
for group in self.all_groups.values():
yield from group.calculations
class Group:
def __init__(self, numbers, all_groups):
self.numbers = numbers
self.size = len(numbers)
self.partitions = list(self.partition_into_unique_pairs(all_groups))
self.calculations = list(self.perform_calculations())
def __repr__(self):
return str(self.numbers)
def partition_into_unique_pairs(self, all_groups):
# The pairs are unordered: a pair (a, b) is equivalent to (b, a).
# Therefore, for pairs of equal length only half of all combinations
# need to be generated to obtain all pairs; this is set by the limit.
if self.size == 1:
return
numbers, size = self.numbers, self.size
limits = (self.halfbinom(size, size//2), )
unique_numbers = set()
for m, limit in zip_longest(range((size+1)//2, size), limits):
for numbers1, numbers2 in self.paired_combinations(numbers, m, limit):
if numbers1 in unique_numbers:
continue
unique_numbers.add(numbers1)
group1, group2 = all_groups[numbers1], all_groups[numbers2]
yield (group1, group2)
def perform_calculations(self):
if self.size == 1:
yield Calculation.singleton(self.numbers[0])
return
for group1, group2 in self.partitions:
for calc1, calc2 in product(group1.calculations, group2.calculations):
yield from Calculation.generate(calc1, calc2)
#classmethod
def paired_combinations(cls, numbers, m, limit):
for cnt, numbers1 in enumerate(combinations(numbers, m), 1):
numbers2 = tuple(cls.filtering(numbers, numbers1))
yield (numbers1, numbers2)
if cnt == limit:
return
#staticmethod
def filtering(iterable, elements):
# filter elements out of an iterable, return the remaining elements
elems = iter(elements)
k = next(elems, None)
for n in iterable:
if n == k:
k = next(elems, None)
else:
yield n
#staticmethod
#lru_cache()
def halfbinom(n, k):
if n % 2 == 1:
return None
prod = 1
for m, l in zip(reversed(range(n+1-k, n+1)), range(1, k+1)):
prod = (prod*m)//l
return prod//2
class Calculation:
def __init__(self, expression, result, is_singleton=False):
self.expr = expression
self.result = result
self.is_singleton = is_singleton
def __repr__(self):
return self.expr
#classmethod
def singleton(cls, n):
return cls(f"{n}", n, is_singleton=True)
#classmethod
def generate(cls, calca, calcb):
if calca.result < calcb.result:
calca, calcb = calcb, calca
for result, op in cls.operations(calca.result, calcb.result):
expr1 = f"{calca.expr}" if calca.is_singleton else f"({calca.expr})"
expr2 = f"{calcb.expr}" if calcb.is_singleton else f"({calcb.expr})"
yield cls(f"{expr1} {op} {expr2}", result)
#staticmethod
def operations(x, y):
yield (x + y, '+')
if x > y: # exclude non-positive results
yield (x - y, '-')
if y > 1 and x > 1: # exclude trivial results
yield (x * y, 'x')
if y > 1 and x % y == 0: # exclude trivial and non-integer results
yield (x // y, '/')
def countdown_solver():
# input: target and numbers. If you want to play with more or less than
# 6 numbers, use the second version of 'unsorted_numbers'.
try:
target = int(sys.argv[1])
unsorted_numbers = (int(sys.argv[n+2]) for n in range(6)) # for 6 numbers
# unsorted_numbers = (int(n) for n in sys.argv[2:]) # for any numbers
numbers = tuple(sorted(unsorted_numbers, reverse=True))
except (IndexError, ValueError):
print("You must provide a target and numbers!")
return
solutions = Solutions(numbers)
smallest_difference = target
bestresults = []
for calculation in solutions.walk():
diff = abs(calculation.result - target)
if diff <= smallest_difference:
if diff < smallest_difference:
bestresults = [calculation]
smallest_difference = diff
else:
bestresults.append(calculation)
output(target, smallest_difference, bestresults)
def output(target, diff, results):
print(f"\nThe closest results differ from {target} by {diff}. They are:\n")
for calculation in results:
print(f"{calculation.result} = {calculation.expr}")
if __name__ == "__main__":
countdown_solver()
The algorithm works as follows:
The numbers are put into a tuple of length 6 in descending order. Then, all unique subgroups of lengths 1 to 6 are created, the smallest groups first.
Example: (75, 50, 5, 9, 1, 1) -> {(75), (50), (9), (5), (1), (75, 50), (75, 9), (75, 5), ..., (75, 50, 9, 5, 1, 1)}.
Next, the groups are organised into a hierarchical tree: every group is partitioned into all unique unordered pairs of its non-empty subgroups.
Example: (9, 5, 1, 1) -> [(9, 5, 1) + (1), (9, 1, 1) + (5), (5, 1, 1) + (9), (9, 5) + (1, 1), (9, 1) + (5, 1)].
Within each group of numbers, the calculations are performed and the results are stored. For groups of length 1, the result is simply the number itself. For larger groups, the calculations are carried out on every pair of subgroups: in each pair, all results of the first subgroup are combined with all results of the second subgroup using +, -, x and /, and the valid outcomes are stored.
Example: (75, 5) consists of the pair ((75), (5)). The result of (75) is 75; the result of (5) is 5; the results of (75, 5) are [75+5=80, 75-5=70, 75*5=375, 75/5=15].
In this manner, all results are generated, from the smallest groups to the largest. Finally, the algorithm iterates through all results and selects the ones that are the closest match to the target number.
For a group of m numbers, the maximum number of arithmetic computations is
comps[m] = 4*sum(binom(m, k)*comps[k]*comps[m-k]//(1 + (2*k)//m) for k in range(1, m//2+1))
For all groups of length 1 to 6, the maximum total number of computations is then
total = sum(binom(n, m)*comps[m] for m in range(1, n+1))
which is 1144386. In practice, it will be much less, because the algorithm reuses the results of duplicate groups, ignores trivial operations (adding 0, multiplying by 1, etc), and because the rules of the game dictate that intermediate results must be positive integers (which limits the use of the division operator).
I think, you need to strictly define the problem first. What you are allowed to do and what you are not. You can start by making it simple and only allowing multiplication, division, substraction and addition.
Now you know your problem space- set of inputs, set of available operations and desired input. If you have only 4 operations and x inputs, the number of combinations is less than:
The number of order in which you can carry out operations (x!) times the possible choices of operations on every step: 4^x. As you can see for 6 numbers it gives reasonable 2949120 operations. This means that this may be your limit for brute force algorithm.
Once you have brute force and you know it works, you can start improving your algorithm with some sort of A* algorithm which would require you to define heuristic functions.
In my opinion the best way to think about it is as the search problem. The main difficulty will be finding good heuristics, or ways to reduce your problem space (if you have numbers that do not add up to the answer, you will need at least one multiplication etc.). Start small, build on that and ask follow up questions once you have some code.
I wrote a terminal application to do this:
https://github.com/pg328/CountdownNumbersGame/tree/main
Inside, I've included an illustration of the calculation of the size of the solution space (it's n*((n-1)!^2)*(2^n-1), so: n=6 -> 2,764,800. I know, gross), and more importantly why that is. My implementation is there if you care to check it out, but in case you don't I'll explain here.
Essentially, at worst it is brute force because as far as I know it's impossible to determine whether any specific branch will result in a valid answer without explicitly checking. Having said that, the average case is some fraction of that; it's {that number} divided by the number of valid solutions (I tend to see around 1000 on my program, where 10 or so are unique and the rest are permutations fo those 10). If I handwaved a number, I'd say roughly 2,765 branches to check which takes like no time. (Yes, even in Python.)
TL;DR: Even though the solution space is huge and it takes a couple million operations to find all solutions, only one answer is needed. Best route is brute force til you find one and spit it out.
I wrote a slightly simpler version:
for every combination of 2 (distinct) elements from the list and combine them using +,-,*,/ (note that since a>b then only a-b is needed and only a/b if a%b=0)
if combination is target then record solution
recursively call on the reduced lists
import sys
def driver():
try:
target = int(sys.argv[1])
nums = list((int(sys.argv[i+2]) for i in range(6)))
except (IndexError, ValueError):
print("Provide a list of 7 numbers")
return
solutions = list()
solve(target, nums, list(), solutions)
unique = set()
final = list()
for s in solutions:
a = '-'.join(sorted(s))
if not a in unique:
unique.add(a)
final.append(s)
for s in final: #print them out
print(s)
def solve(target, nums, path, solutions):
if len(nums) == 1:
return
distinct = sorted(list(set(nums)), reverse = True)
rem1 = list(distinct)
for n1 in distinct: #reduce list by combining a pair
rem1.remove(n1)
for n2 in rem1:
rem2 = list(nums) # in case of duplicates we need to start with full list and take out the n1,n2 pair of elements
rem2.remove(n1)
rem2.remove(n2)
combine(target, solutions, path, rem2, n1, n2, '+')
combine(target, solutions, path, rem2, n1, n2, '-')
if n2 > 1:
combine(target, solutions, path, rem2, n1, n2, '*')
if not n1 % n2:
combine(target, solutions, path, rem2, n1, n2, '//')
def combine(target, solutions, path, rem2, n1, n2, symb):
lst = list(rem2)
ans = eval("{0}{2}{1}".format(n1, n2, symb))
newpath = path + ["{0}{3}{1}={2}".format(n1, n2, ans, symb[0])]
if ans == target:
solutions += [newpath]
else:
lst.append(ans)
solve(target, lst, newpath, solutions)
if __name__ == "__main__":
driver()
I came across another codechef problem which I am attempting to solve in Scala. The problem statement is as follows:
Stepford Street was a dead end street. The houses on Stepford Street
were bought by wealthy millionaires. They had them extensively altered
so that as one progressed along the street, the height of the
buildings increased rapidly. However, not all millionaires were
created equal. Some refused to follow this trend and kept their houses
at their original heights. The resulting progression of heights was
thus disturbed. A contest to locate the most ordered street was
announced by the Beverly Hills Municipal Corporation. The criteria for
the most ordered street was set as follows: If there exists a house
with a lower height later in the street than the house under
consideration, then the pair (current house, later house) counts as 1
point towards the disorderliness index of the street. It is not
necessary that the later house be adjacent to the current house. Note:
No two houses on a street will be of the same height For example, for
the input: 1 2 4 5 3 6 The pairs (4,3), (5,3) form disordered pairs.
Thus the disorderliness index of this array is 2. As the criteria for
determining the disorderliness is complex, the BHMC has requested your
help to automate the process. You need to write an efficient program
that calculates the disorderliness index of a street.
A sample input output provided is as follows:
Input: 1 2 4 5 3 6
Output: 2
The output is 2 because of two pairs (4,3) and (5,3)
To solve this problem I thought I should use a variant of MergeSort,incrementing by 1 when the left element is greater than the right element.
My scala code is as follows:
def dysfunctionCalc(input:List[Int]):Int = {
val leftHalf = input.size/2
println("HalfSize:"+leftHalf)
val isOdd = input.size%2
println("Is odd:"+isOdd)
val leftList = input.take(leftHalf+isOdd)
println("LeftList:"+leftList)
val rightList = input.drop(leftHalf+isOdd)
println("RightList:"+rightList)
if ((leftList.size <= 1) && (rightList.size <= 1)){
println("Entering input where both lists are <= 1")
if(leftList.size == 0 || rightList.size == 0){
println("One of the lists is less than 0")
0
}
else if(leftList.head > rightList.head)1 else 0
}
else{
println("Both lists are greater than 1")
dysfunctionCalc(leftList) + dysfunctionCalc(rightList)
}
}
First off, my logic is wrong,it doesn't have a merge stage and I am not sure what would be the best way to percolate the result of the base-case up the stack and compare it with the other values. Also, using recursion to solve this problem may not be the most optimal way to go since for large lists, I maybe blowing up the stack. Also, there might be stylistic issues with my code as well.
I would be great if somebody could point out other flaws and the right way to solve this problem.
Thanks
Suppose you split your list into three pieces: the item you are considering, those on the left, and those on the right. Suppose further that those on the left are in a sorted set. Now you just need to walk through the list, moving items from "right" to "considered" and from "considered" to "left"; at each point, you look at the size of the subset of the sorted set that is greater than your item. In general, the size lookup can be done in O(log(N)) as can the add-element (with a Red-Black or AVL tree, for instance). So you have O(N log N) performance.
Now the question is how to implement this in Scala efficiently. It turns out that Scala has a Red-Black tree used for its TreeSet sorted set, and the implementation is actually quite simple (here in tail-recursive form):
import collection.immutable.TreeSet
final def calcDisorder(xs: List[Int], left: TreeSet[Int] = TreeSet.empty, n: Int = 0): Int = xs match {
case Nil => n
case x :: rest => calcDisorder(rest, left + x, n + left.from(x).size)
}
Unfortunately, left.from(x).size takes O(N) time (I believe), which yields a quadratic execution time. That's no good--what you need is an IndexedTreeSet which can do indexOf(x) in O(log(n)) (and then iterate with n + left.size - left.indexOf(x) - 1). You can build your own implementation or find one on the web. For instance, I found one here (API here) for Java that does exactly the right thing.
Incidentally, the problem with doing a mergesort is that you cannot easily work cumulatively. With merging a pair, you can keep track of how out-of-order it is. But when you merge in a third list, you must see how out of order it is with respect to both other lists, which spoils your divide-and-conquer strategy. (I am not sure whether there is some invariant one could find that would allow you to calculate directly if you kept track of it.)
Here is my try, I don't use MergeSort but it seems to solve the problem:
def calcDisorderness(myList:List[Int]):Int = myList match{
case Nil => 0
case t::q => q.count(_<t) + calcDisorderness(q)
}
scala> val input = List(1,2,4,5,3,6)
input: List[Int] = List(1, 2, 4, 5, 3, 6)
scala> calcDisorderness(input)
res1: Int = 2
The question is, is there a way to have a lower complexity?
Edit: tail recursive version of the same function and cool usage of default values in function arguments.
def calcDisorderness(myList:List[Int], disorder:Int=0):Int = myList match{
case Nil => disorder
case t::q => calcDisorderness(q, disorder + q.count(_<t))
}
A solution based on Merge Sort. Not super fast, potential slowdown could be in "xs.length".
def countSwaps(a: Array[Int]): Long = {
var disorder: Long = 0
def msort(xs: List[Int]): List[Int] = {
import Stream._
def merge(left: List[Int], right: List[Int], inc: Int): Stream[Int] = {
(left, right) match {
case (x :: xs, y :: ys) if x > y =>
cons(y, merge(left, ys, inc + 1))
case (x :: xs, _) => {
disorder += inc
cons(x, merge(xs, right, inc))
}
case _ => right.toStream
}
}
val n = xs.length / 2
if (n == 0)
xs
else {
val (ys, zs) = xs splitAt n
merge(msort(ys), msort(zs), 0).toList
}
}
msort(a.toList)
disorder
}
Another solution based on Merge Sort. Very fast: no FP or for-loop.
def countSwaps(a: Array[Int]): Count = {
var swaps: Count = 0
def mergeRun(begin: Int, run_len: Int, src: Array[Int], dst: Array[Int]) = {
var li = begin
val lend = math.min(begin + run_len, src.length)
var ri = begin + run_len
val rend = math.min(begin + run_len * 2, src.length)
var ti = begin
while (ti < rend) {
if (ri >= rend) {
dst(ti) = src(li); li += 1
swaps += ri - begin - run_len
} else if (li >= lend) {
dst(ti) = src(ri); ri += 1
} else if (a(li) <= a(ri)) {
dst(ti) = src(li); li += 1
swaps += ri - begin - run_len
} else {
dst(ti) = src(ri); ri += 1
}
ti += 1
}
}
val b = new Array[Int](a.length)
var run = 0
var run_len = 1
while (run_len < a.length) {
var begin = 0
while (begin < a.length) {
val (src, dst) = if (run % 2 == 0) (a, b) else (b, a)
mergeRun(begin, run_len, src, dst)
begin += run_len * 2
}
run += 1
run_len *= 2
}
swaps
}
Convert the above code to Functional style: no mutable variable, no loop.
All recursions are tail calls, thus the performance is good.
def countSwaps(a: Array[Int]): Count = {
def mergeRun(li: Int, lend: Int, rb: Int, ri: Int, rend: Int, di: Int, src: Array[Int], dst: Array[Int], swaps: Count): Count = {
if (ri >= rend && li >= lend) {
swaps
} else if (ri >= rend) {
dst(di) = src(li)
mergeRun(li + 1, lend, rb, ri, rend, di + 1, src, dst, ri - rb + swaps)
} else if (li >= lend) {
dst(di) = src(ri)
mergeRun(li, lend, rb, ri + 1, rend, di + 1, src, dst, swaps)
} else if (src(li) <= src(ri)) {
dst(di) = src(li)
mergeRun(li + 1, lend, rb, ri, rend, di + 1, src, dst, ri - rb + swaps)
} else {
dst(di) = src(ri)
mergeRun(li, lend, rb, ri + 1, rend, di + 1, src, dst, swaps)
}
}
val b = new Array[Int](a.length)
def merge(run: Int, run_len: Int, lb: Int, swaps: Count): Count = {
if (run_len >= a.length) {
swaps
} else if (lb >= a.length) {
merge(run + 1, run_len * 2, 0, swaps)
} else {
val lend = math.min(lb + run_len, a.length)
val rb = lb + run_len
val rend = math.min(rb + run_len, a.length)
val (src, dst) = if (run % 2 == 0) (a, b) else (b, a)
val inc_swaps = mergeRun(lb, lend, rb, rb, rend, lb, src, dst, 0)
merge(run, run_len, lb + run_len * 2, inc_swaps + swaps)
}
}
merge(0, 1, 0, 0)
}
It seems to me that the key is to break the list into a series of ascending sequences. For example, your example would be broken into (1 2 4 5)(3 6). None of the items in the first list can end a pair. Now you do a kind of merge of these two lists, working backwards:
6 > 5, so 6 can't be in any pairs
3 < 5, so its a pair
3 < 4, so its a pair
3 > 2, so we're done
I'm not clear from the definition on how to handle more than 2 such sequences.