We Keep Coding

Why is my recursive memoized Fibonacci code wrong?

Something in the main function is wrong, but I don't know what. If I go through the code with a debugger I can see that my code isn't even reaching the fibonacci function right now.
public class Fibonacci {
public static void main(String[] args) {
for (int n=1; n<50; n++) {
System.out.println("Element "+ n + " of the sequence: " + newFib(n));
}
}
public static ArrayList<BigInteger> memo = new ArrayList<BigInteger>();
static BigInteger newFib(int n){
assert n >= 1: "the fibonacci sequence starts at 1";
BigInteger result=BigInteger.valueOf(1);
if (memo.get(n) != null) {
return memo.get(n);
}
else if( n == 1 || n == 2 ) {
memo.add(n, BigInteger.valueOf(1));
return BigInteger.valueOf(1);
}
else {
result= newFib(n-1).add(newFib(n-2));
memo.add(n,result);
return result;
}
}
}

Your code was throwing some exceptions. You simply needed to debug the exceptions and implement the proper checks. Using the code with these changes should work as intended:
import java.math.*;
import java.util.ArrayList;
public class Fibonacci {
public static void main(String[] args) {
for (int n=1; n<50; n++) {
System.out.println("Element "+ n + " of the sequence: " + newFib(n));
}
}
public static ArrayList<BigInteger> memo = new ArrayList<BigInteger>();
static BigInteger newFib(int n){
assert n >= 1: "the fibonacci sequence starts at 1";
BigInteger result=BigInteger.valueOf(1);
if (memo.size() - 1 >= n && memo.get(n) != null) {
return memo.get(n);
}
else if( n == 1 || n == 2 ) {
memo.add(n-1, BigInteger.valueOf(1));
return BigInteger.valueOf(1);
}
else {
result= newFib(n-1).add(newFib(n-2));
memo.add(n,result);
return result;
}
}
}

Your current code does not handle out of bounds correctly and as I mentioned in the comments, the fibonacci sequence does not necessarily start at one and the logic can be simplified. I would use a HashMap<Integer, BigInteger> for storing the memoization and I would also prefer to populate the initial constants once (and as constants). For example,
private static Map<Integer, BigInteger> memo = new HashMap<>();
static {
memo.put(0, BigInteger.ZERO);
memo.put(1, BigInteger.ONE);
}
static BigInteger newFib(int n) {
if (!memo.containsKey(n)) {
if (n < 0) {
memo.put(n, newFib(n + 2).subtract(newFib(n + 1)));
} else {
memo.put(n, newFib(n - 2).add(newFib(n - 1)));
}
}
return memo.get(n);
}
Which can then be tested like
public static void main(String[] args) {
for (int i = -8; i < 9; i++) {
if (i != -8) {
System.out.print(" ");
}
System.out.printf("%-5s", String.format("F(%d)", i));
}
System.out.println();
for (int i = -8; i < 9; i++) {
if (i != -8) {
System.out.print(" ");
}
System.out.printf("%-5s", newFib(i));
}
}
To reproduce the example given in the Fibonacci number Wikipedia entry.
F(-8) F(-7) F(-6) F(-5) F(-4) F(-3) F(-2) F(-1) F(0) F(1) F(2) F(3) F(4) F(5) F(6) F(7) F(8)
-21 13 -8 5 -3 2 -1 1 0 1 1 2 3 5 8 13 21

Forming a pattern of bits from a integer

I need help to design java code for generating bit array for any given integer in following manner:
23 should produce output as 1101011 (min length array)
explaination :
positions are given as 1 -2 4 -8 16 -32 ....
So 1101011 can be evaluated as:
1*1 + 1*-2 + 0*4+ 1*-8 + 0*16 +1*-32 + 1*64 = 23

This is the so-called negabinary representation of numbers (described first by Vittorio Grünwald in 1885). They can be encoded in a fashion very similar to the usual binary representation, just working with -2 instead of 2 as base (Java code inspired by C# code on https://en.wikipedia.org/wiki/Negative_base ):
class EncodeNegaBinary {
public static void main(String[] args) {
int n=0,input=0;
String result="";
final String[] BITS = { "0","1" };
if (args.length != 1) {
System.err.println("Please enter an integer to be converted");
return;
} else {
input = n = Integer.parseInt(args[0]);
}
while (n != 0) {
int r = n%-2;
n /= -2;
if (r == -1) {
r=1;
n++;
}
result = BITS[r] + result;
}
System.out.printf( "%d -> %s\n", input, result);
}
}

Since it is not usual int to binary conversion, at each step we need to consider two cases as at each position there can be only two choices 0 or 1. This is done recursively in the below program:
public class ModifiedIntToBinaryConversion{
public static int calcBinaryString(int reqSum, int currSum, int add, String bs) {
if (reqSum == currSum) { // base condtion 1
System.out.println("The string is \n" + bs);
return currSum;
}
if (add + currSum > reqSum) { // base condtion 2
return 0;
}
int newAdd = add * -2;
// System.out.println("new add is "+ newAdd +" currSum is "+ currSum);
int s1 = calcBinaryString(reqSum, currSum + add, newAdd, bs + "1");
if (s1 == reqSum)
return s1;
int s2 = calcBinaryString(reqSum, currSum, newAdd, bs + "0");
return s2;
}
public static void calcBinaryString(int sum) {
int s1 = calcBinaryString(sum, 0, 1, "");
if(s1 != sum) {
System.out.println("The binary equivalent couldn't be found");
}
}
public static void main(String[] args) {
calcBinaryString(23);
}
}
Now base condition 1 is clear as I am just checking whether required sum and calculated sum are equal.
For base condition 2, I will accept it's result of debugging and a bit of thought as I was getting Stackoverflow errors. Once the calculated sum becomes greater than the required sum and then we take the next -ve number so that it become less than req. sum. But then the next +ve number will be greater than the -ve number we just considered and thus the chances are very less that the calculated sum will ever be equal to req. sum.

Creating a Complex number class in Java

I am a new in java and programming in general.
I am currently doing complex numbers. I know that there might be an answer for this online, but it will also reveal if I used the correct algorithm and so on, so I am trying to avoid other answers around here.
My main issue is that I am having trouble with the Divide class that I made, since in complex number division we are going to return a fraction as an answer, I can't figure out how to have the program return the 2 statements that it calculate, can someone advise what can be done? Here is the part of the division code, it works fine when I check both part 1 and then part 2, but how can i get it to return both of them when calling using that class.
I attached my full code that I made, I know it can be tweaked to have less coding, but that is not my current concern.
Thanks for your help in advance.
class complexN {
int R, I;
complexN(int R, int I) {
this.R = R;
this.I = I;
}
complexN AddCN(complexN A) {
return new complexN(this.R + A.R, this.I + A.I);
}
complexN SubCN(complexN A) {
int AnsR = this.R - A.R;
int AnsI = this.I - A.I;
complexN Sum = new complexN(AnsR, AnsI);
return Sum;
}
complexN MulCN(complexN A) {
int AnsI = (this.R * A.I) + (this.I * A.R);
int AnsR = (this.R * A.R) - (this.I * A.I);
complexN Sum = new complexN(AnsR, AnsI);
return Sum;
}
complexN DivCN(complexN A) {
complexN ComCon = new complexN(A.R, (A.I * -1));
complexN part1 = new complexN(this.R, this.I).MulCN(ComCon);
complexN part2 = A.MulCN(ComCon);
return part1;
}
void print() {
String i = (this.I == 1 ? "" : (this.I == -1 ? "-" : "" + this.I));
if (this.R != 0 && this.I > 0) {
System.out.println(this.R + "+" + i + "i");
}
if (this.R != 0 && this.I < 0) {
System.out.println(this.R + i + "i");
}
if (this.R != 0 && this.I == 0) {
System.out.println(this.R);
}
if (this.R == 0 && this.I != 0) {
System.out.println(i + "i");
}
if (this.R == 0 && this.I == 0) {
System.out.println("0");
}
}
}
class complex {
public static void main(String[] args) {
// TODO Auto-generated method stub
complexN z1 = new complexN(5, 2);
complexN z2 = new complexN(3, -4);
System.out.print("z1 = ");
z1.print();
System.out.print("z2 = ");
z2.print();
System.out.println("---------");
z1.DivCN(z2).print();
}
}

How to pass a String parameter to a void method? [duplicate]

This question already has an answer here:
What does "Incompatible types: void cannot be converted to ..." mean?
(1 answer)
Closed 3 years ago.
I need to write a Java program for a course I'm taking which looks for genes in a strand of DNA.The Issue I am having is that from the test method, I need to pass printAllgenes(a) to the void printAllgenes method. In the test method I've tried setting 'int a' to 'String a', but in either case an error when compiling explaining that void cannot be converted to int or String. I'm sure its obvious, but I'm very new to programming, so please pardon my ignorance! Thank you.
import java.io.*;
import edu.duke.*;
public class FindProtein {
public void test() {
String a = "atg aaa tab tag atg aaa tga aat ag";
int b = printAllgenes(a);
System.out.println("DNA string is " + a);
System.out.println("Gene found is " + b);
}
public void printAllgenes(String dna) {
int sp = 0; //start point
while (true) {
int start = dna.indexOf("atg,sp");
if (start == -1) {
break;
}
int stop = findStopIndex(dna, start + 3);
if (stop != dna.length()) {
System.out.println(dna.substring(start, stop + 3));
sp = stop + 3;
} else {
sp = sp + 3;
}
}
}
public int findStopIndex(String dna, int index) {
int tga = dna.indexOf("tga", index);
if (tga == -1 || (tga - index) % 3 != 0) {
tga = dna.length();
}
int taa = dna.indexOf("taa", index);
if (taa == -1 || (taa - index) % 3 != 0) {
taa = dna.length();
}
int tag = dna.indexOf("tag", index);
if (tag == -1 || (tga - index) % 3 != 0) {
tag = dna.length();
}
return Math.min(tga, Math.min(taa, tag));
}
}

Try to use just:
printAllgenes(a);
Because printAllgenes method doesn't have any type of return statement.

change return type void to int It will return your count whatever u want to return from printAllgenes(String dns) Method. You will get a int return which will initialize you variable b that is being displayed on Console.
public int printAllgenes(String dna){
int sp = 0; //start point
while (true){
int start = dna.indexOf("atg,sp");
if (start==-1){
break;
}
int stop = findStopIndex(dna,start+3);
if (stop!=dna.length()){
System.out.println(dna.substring(start,stop+3));
sp=stop+3;
}
else{
sp=sp+3;
}
}
return sp;
}
Now Your Test Method Implementation will work fine...
public void test(){
String a= "atg aaa tab tag atg aaa tga aat ag";
int b = printAllgenes(a);
System.out.println("DNA string is " +a);
System.out.println("Gene found is "+b);
}
Thank you..

Pseudo first class functions in Java?

Below I have a rough implementation of testing a list lambda functions on an integer in Python. I know Java currently doesn't support closures or lambdas (yet), but I'm curious, is this remotely possible in Java?
pos = [lambda x : x > 0, "pos"]
even = [lambda x : x % 2 == 0, "even"]
small = [lambda x : x < 100, "small"]
def pass_test(x, tests):
if (len(tests) == 0): return []
else:
if (tests[0][0](x)): return [tests[0][1]] + pass_test(x, tests[1:])
else: return pass_test(x, tests[1:])

Yes, it is possible, along the lines of:
interface Predicate<A> {
boolean eval(A argument);
}
Predicate<Integer> gt0 = new Predicate<Integer>() {
boolean eval(Integer arg) {
return arg > 0;
}
};
You see, this is a bit verbose, but it does the job.

In your case, python function could be mapped to Java like this:
public interface Function {
String do(int x);
}
// ...
Function pos = new Function(){
public String do(int x) {
return (x > 0) ? "pos" : "";
}
}
Function even = new Function(){
public String do(int x) {
return (x % 2 == 0) ? "even" : "";
}
}
Function small = new Function(){
public String do(int x) {
return (x < 100) ? "small" : "";
}
}
// ...
As you see, you'd need a lot more code to do the same simple thing in Java.