Python has a really sophisticated way of handling iterations. The only thing it does not have “GOTO Labels” which I think is good.<\/p>\n\n\n\n
Let us compare the three common ways of iterations in Python: While, For and Map by the way of an example. Imagine that you have a list of numbers and you would like to find the square of each number.<\/p>\n\n\n\n
nums = [1,2,3,5,10]\nresult = []\nfor num in nums:\n result.append(num*num)\nprint(result)\n<\/code><\/pre>\n\n\n\nIt would print [1, 4, 9, 25, 100]<\/code><\/p>\n\n\n\n\n\n\n\nAlternative form of iteration using for is <\/p>\n\n\n\n
nums = [1,2,3,5,10]\nresult = [num*num for num in nums]\nprint(result)<\/code><\/pre>\n\n\n\nNow, let us try to use while<\/code>.<\/p>\n\n\n\nnums = [1,2,3,5,10]\nresult = []\ni = 0\nwhile i < len(nums):\n num = nums[i]\n result.append(num*num)\n i += 1\n\nprint(result)<\/code><\/pre>\n\n\n\nLook at both pieces of code. The for<\/code> loop iterates over a list and while<\/code> loop runs as long as the condition is true. So, if you forgot to increase i<\/code>, the condition will remain false forever and while<\/code> the loop will not end ever. That’s why while<\/code> is called an indefinite loop. <\/p>\n\n\n\nNot everything that can be achieved using while<\/code> can be achieved using for<\/code>. Everything that can be achieved using for<\/code> can be achieved using while<\/code>. while<\/code> is way more powerful but with more power comes more risk.<\/p>\n\n\n\nSo, if your task can be performed using for<\/code> don’t use while<\/code>.<\/p>\n\n\n\nNow, let’s take a look at map<\/code>. The map is a very interesting way of running the independent operations on a list of things. I encountered the map paradigm around 18 years ago while doing grid computing using PERL programming language.<\/p>\n\n\n\nLet’s take a look at the code to solve the same problem as above – squaring each number in a list.<\/p>\n\n\n\n
nums = [1,2,3,5,10]\ndef sq(x):\n return x*x\n\nresult = map(sq, nums)\nprint(result)<\/code><\/pre>\n\n\n\n Here notice that to the function we are passing our function sq<\/code> and our arraynums<\/code>. The map<\/code> function will run the sq<\/code> function on each element of nums. Here it does not matter to us in what order will it run. Also, whether it uses different processors to run sq<\/code> function on different parts of the list it would not impact our logic.<\/p>\n\n\n\nThe map<\/code> function is way better to use over for<\/code> loop but it has constraint. If the next value depends upon the previous computation, we would not be able to use map. So, map<\/code> can only solve some problems that for<\/code> can solve. And for<\/code> can solve some problems that while<\/code> can.<\/p>\n\n\n\nAdvantage of map<\/code> is that your code will easily be parallelizable i.e. run on multiple computers or multiple processors if you are using map<\/code>. The code with for<\/code> loop and while<\/code> loops we will have to rewrite the logic to make it parallelizable.<\/p>\n\n\n\nLet us take the case of the Fibonacci series. Can we achieve the following using map()? I don’t think so. If you can, please put it in the comments.<\/p>\n\n\n\n
def fib(n):\n result = []\n first, second = 0, 1\n result.extend([first, second])\n for _ in range(n-1):\n newnum = first + second\n first, second = second, newnum\n result.append(newnum)\n print(result)<\/code><\/pre>\n\n\n\n>>> fib(20)\n[0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987, 1597, 2584, 4181, 6765] <\/pre>\n\n\n\nThis prints first twenty numbers of Fibonacci<\/code> series.<\/p>\n\n\n\nThere are many examples which are very difficult to achieve using map. <\/p>\n\n\n\n
As a conclusion, your order of preference of usage should be:<\/p>\n\n\n\n
map > for > while<\/strong><\/code><\/pre>\n\n\n\nPrefer map<\/code> over “for” loop. Prefer for<\/code> over “while” loop.<\/p>\n\n\n\nI hope this discussion was useful to you. You can try it on CloudxLab<\/a> right away.<\/p>\n\n\n\nHappy Learning!<\/p>\n","protected":false},"excerpt":{"rendered":"