Python ile veri yap\u0131lar\u0131 ve algoritmalar\u0131 \u00f6\u011frenmek hi\u00e7 bu kadar kolay olmam\u0131\u015ft\u0131! Bu kapsaml\u0131 rehber, DSA m\u00fclakatlar\u0131na haz\u0131rlan\u0131rken ve projelerinizde en iyi performans\u0131 ararken size yol g\u00f6sterecek.<\/p>\n
Modern yaz\u0131l\u0131m geli\u015ftirme d\u00fcnyas\u0131nda, \u00f6zellikle de Python gibi y\u00fcksek seviyeli ve \u00e7ok y\u00f6nl\u00fc bir dille \u00e7al\u0131\u015f\u0131yorsan\u0131z, veri yap\u0131lar\u0131 ve algoritmalar (DSA) konusundaki yetkinli\u011finiz, sadece teorik bilginin \u00f6tesinde bir rekabet avantaj\u0131 sa\u011flar. Peki, neden bu kadar \u00f6nemli? Diyelim ki bir e-ticaret platformunda \u00e7al\u0131\u015f\u0131yorsunuz ve milyonlarca \u00fcr\u00fcn verisiyle u\u011fra\u015f\u0131yorsunuz. M\u00fc\u015fterilerin arama sorgular\u0131na saniyeler i\u00e7inde yan\u0131t vermek, stok seviyelerini anl\u0131k olarak g\u00fcncellemek ve \u00f6neri sistemlerini verimli bir \u015fekilde \u00e7al\u0131\u015ft\u0131rmak zorundas\u0131n\u0131z. \u0130\u015fte bu noktada, do\u011fru veri yap\u0131s\u0131n\u0131 se\u00e7mek ve en uygun algoritmay\u0131 uygulamak, uygulaman\u0131z\u0131n performans\u0131n\u0131, \u00f6l\u00e7eklenebilirli\u011fini ve kullan\u0131c\u0131 deneyimini do\u011frudan etkiler.<\/p>\n
Yanl\u0131\u015f bir veri yap\u0131s\u0131 se\u00e7imi, basit bir arama i\u015fleminin bile uygulaman\u0131z\u0131 yava\u015flatmas\u0131na veya a\u015f\u0131r\u0131 bellek t\u00fcketmesine neden olabilir. \u00d6rne\u011fin, s\u0131kl\u0131kla eleman ekleyip \u00e7\u0131kard\u0131\u011f\u0131n\u0131z ve ortadan eri\u015fim yapt\u0131\u011f\u0131n\u0131z bir senaryoda ba\u011fl\u0131 liste yerine dinamik bir dizi (Python listesi) kullanmak, her operasyonda gereksiz bellek kopyalamalar\u0131na ve performans d\u00fc\u015f\u00fc\u015flerine yol a\u00e7abilir. Bu nedenle, sadece kod yazmay\u0131 bilmek yeterli de\u011fildir; ayn\u0131 zamanda yazd\u0131\u011f\u0131n\u0131z kodun nas\u0131l \u00e7al\u0131\u015ft\u0131\u011f\u0131n\u0131, ne kadar kaynak t\u00fcketti\u011fini ve farkl\u0131 senaryolarda nas\u0131l performans g\u00f6sterece\u011fini de anlaman\u0131z gerekir. Bu anlay\u0131\u015f, sizi basit bir kod yaz\u0131c\u0131s\u0131ndan, problemleri verimli bir \u015fekilde \u00e7\u00f6zen ve \u00f6l\u00e7eklenebilir sistemler tasarlayan bir m\u00fchendise d\u00f6n\u00fc\u015ft\u00fcr\u00fcr. Python’\u0131n esnekli\u011fi ve zengin k\u00fct\u00fcphane ekosistemi, DSA’y\u0131 \u00f6\u011frenmeyi ve uygulamay\u0131 son derece eri\u015filebilir k\u0131lar, ancak bu ara\u00e7lar\u0131 etkili kullanabilmek i\u00e7in temel prensipleri kavramak \u015fartt\u0131r.<\/p>\n
Veri yap\u0131lar\u0131 ve algoritmalar, bilgisayar bilimlerinin temel ta\u015flar\u0131d\u0131r. Veri yap\u0131lar\u0131, verileri bilgisayar belle\u011finde d\u00fczenlemenin ve saklaman\u0131n belirli yollar\u0131yken, algoritmalar bu veriler \u00fczerinde belirli g\u00f6revleri yerine getirmek i\u00e7in ad\u0131m ad\u0131m talimatlar dizisidir. Bu iki kavram birbiriyle ayr\u0131lmaz bir b\u00fct\u00fcnd\u00fcr; \u00e7\u00fcnk\u00fc bir algoritman\u0131n verimli \u00e7al\u0131\u015fmas\u0131, genellikle \u00fczerinde \u00e7al\u0131\u015ft\u0131\u011f\u0131 verilerin nas\u0131l yap\u0131land\u0131r\u0131ld\u0131\u011f\u0131na ba\u011fl\u0131d\u0131r. Ba\u015flang\u0131\u00e7 seviyesindeki bir geli\u015ftirici i\u00e7in bu konular biraz soyut gelebilir, ancak g\u00fcnl\u00fck hayattaki kar\u015f\u0131l\u0131klar\u0131n\u0131 d\u00fc\u015f\u00fcnd\u00fc\u011f\u00fcm\u00fczde anlamas\u0131 kolayla\u015f\u0131r. \u00d6rne\u011fin, bir k\u00fct\u00fcphanede kitaplar\u0131 nas\u0131l d\u00fczenledi\u011finiz bir veri yap\u0131s\u0131d\u0131r; kitaplar\u0131 bulmak i\u00e7in uygulad\u0131\u011f\u0131n\u0131z y\u00f6ntem (alfabetik s\u0131raya g\u00f6re aramak gibi) ise bir algoritmad\u0131r.<\/p>\n
Veri yap\u0131lar\u0131, verileri bilgisayar belle\u011finde depolaman\u0131n, organize etmenin ve bunlara eri\u015fmenin belirli yollar\u0131d\u0131r. Amac\u0131, belirli bir ama\u00e7 i\u00e7in veri depolama ve eri\u015fimi optimize etmektir. Bir veri yap\u0131s\u0131 se\u00e7imi, bir uygulaman\u0131n verimlili\u011fini, h\u0131z\u0131n\u0131 ve bellek kullan\u0131m\u0131n\u0131 do\u011frudan etkiler. Yanl\u0131\u015f bir se\u00e7im, gereksiz hesaplama s\u00fcresine veya bellek israf\u0131na yol a\u00e7abilir. \u00d6rne\u011fin, sabit boyutlu bir koleksiyonda h\u0131zl\u0131 eri\u015fim gerektiren durumlarda diziler (listeler), s\u0131k s\u0131k eleman ekleme ve silme i\u015flemlerinin oldu\u011fu ancak s\u0131ral\u0131 eri\u015fimin az oldu\u011fu durumlarda ba\u011fl\u0131 listeler daha uygun olabilir. Veri yap\u0131lar\u0131 sadece depolama mekanizmalar\u0131 de\u011fildir; ayn\u0131 zamanda belirli operasyonlar (arama, ekleme, silme, g\u00fcncelleme) i\u00e7in farkl\u0131 performans karakterleri sunarlar. Bu karakterleri anlamak, kar\u015f\u0131la\u015ft\u0131\u011f\u0131n\u0131z problemi en verimli \u015fekilde \u00e7\u00f6zmenizi sa\u011flar.<\/p>\n
Algoritma, belirli bir problemi \u00e7\u00f6zmek veya belirli bir g\u00f6revi yerine getirmek i\u00e7in tan\u0131mlanm\u0131\u015f, ad\u0131m ad\u0131m talimatlar dizisidir. Algoritmalar, bir input al\u0131r, belirli bir i\u015flem dizisini uygular ve bir output \u00fcretir. \u00d6rne\u011fin, bir say\u0131 listesini s\u0131ralamak (input), belirli bir s\u0131ralama algoritmas\u0131n\u0131 (i\u015flem dizisi) uygulamak ve s\u0131ral\u0131 listeyi (output) elde etmek bir algoritmad\u0131r. Bir algoritman\u0131n “ba\u015far\u0131l\u0131” kabul edilmesi i\u00e7in sadece do\u011fru sonu\u00e7 \u00fcretmesi yetmez; ayn\u0131 zamanda verimli olmas\u0131, yani minimum zaman ve bellek kaynaklar\u0131 kullanarak \u00e7al\u0131\u015fmas\u0131 gerekir. Bu verimlilik genellikle “Big O Notasyonu” ile ifade edilir, ki bu konuyu ileri d\u00fczey b\u00f6l\u00fcmde detayl\u0131ca ele alaca\u011f\u0131z. Ba\u015far\u0131l\u0131 bir algoritma uygulamas\u0131 i\u00e7in anahtar, problemi iyi anlamak, farkl\u0131 \u00e7\u00f6z\u00fcm yakla\u015f\u0131mlar\u0131n\u0131 de\u011ferlendirmek ve en uygun algoritmay\u0131 se\u00e7mektir. Bu, genellikle deneme-yan\u0131lma, problem par\u00e7alama ve soyutlama becerisi gerektiren bir s\u00fcre\u00e7tir.<\/p>\n
Python, veri yap\u0131lar\u0131 ve algoritmalar\u0131 \u00f6\u011frenmek ve uygulamak i\u00e7in harika bir dildir. Bunun birka\u00e7 temel nedeni vard\u0131r:<\/p>\n
collections<\/code> (deque<\/code>, Counter<\/code>), heapq<\/code> (min-heap), functools<\/code> gibi mod\u00fcller, daha karma\u015f\u0131k veri yap\u0131lar\u0131 ve algoritmalar i\u00e7in haz\u0131r ara\u00e7lar sunar.<\/li>\n- H\u0131zl\u0131 Prototipleme:<\/b> Python’\u0131n yorumlay\u0131c\u0131 tabanl\u0131 yap\u0131s\u0131 ve dinamik yaz\u0131m\u0131, fikirleri ve algoritmalar\u0131 h\u0131zla test etmenize ve prototiplere d\u00f6n\u00fc\u015ft\u00fcrmenize olanak tan\u0131r.<\/li>\n<\/ul>\n
Bu \u00f6zellikler, Python’\u0131 hem teorik DSA kavramlar\u0131n\u0131 anlamak hem de bu kavramlar\u0131 ger\u00e7ek d\u00fcnya problemlerine uygulamak i\u00e7in ideal bir ara\u00e7 haline getirir.<\/p>\n
Uygulamal\u0131 K\u0131s\u0131m: Temel Veri Yap\u0131lar\u0131 Python ile Nas\u0131l Uygulan\u0131r?<\/h2>\n
\u015eimdi s\u0131ra geldi teorik bilgileri prati\u011fe d\u00f6kmeye. Python’\u0131n yerle\u015fik yeteneklerini kullanarak en yayg\u0131n veri yap\u0131lar\u0131n\u0131 nas\u0131l olu\u015fturaca\u011f\u0131m\u0131za ve y\u00f6netece\u011fimize bakal\u0131m. Bu b\u00f6l\u00fcm, her bir veri yap\u0131s\u0131n\u0131n temel mant\u0131\u011f\u0131n\u0131, Python’daki kar\u015f\u0131l\u0131\u011f\u0131n\u0131 ve yayg\u0131n kullan\u0131m senaryolar\u0131n\u0131 ad\u0131m ad\u0131m a\u00e7\u0131klayacakt\u0131r. Unutmay\u0131n, do\u011fru veri yap\u0131s\u0131 se\u00e7imi, algoritman\u0131z\u0131n genel performans\u0131n\u0131 belirleyen kritik bir ad\u0131md\u0131r.<\/p>\n
Diziler (Listeler) ve Temel \u0130\u015flemler: H\u0131zl\u0131 Eri\u015fim \u0130\u00e7in Ne Bilmelisiniz?<\/h3>\n
Python’da dizilerin kar\u015f\u0131l\u0131\u011f\u0131 <\/p>\n
list<\/pre>\n<\/code> veri yap\u0131s\u0131d\u0131r. Python listeleri dinamiktir, yani boyutlar\u0131 \u00e7al\u0131\u015fma zaman\u0131nda de\u011fi\u015febilir. Bellekte biti\u015fik bir blok olarak tutulur ve elemanlara indeksleri arac\u0131l\u0131\u011f\u0131yla <\/p>\n
O(1)<\/pre>\n<\/code> zaman karma\u015f\u0131kl\u0131\u011f\u0131nda eri\u015filebilir. Ancak, listenin ba\u015f\u0131na veya ortas\u0131na eleman eklemek\/\u00e7\u0131karmak <\/p>\n
O(n)<\/pre>\n<\/code> zaman alabilir, \u00e7\u00fcnk\u00fc di\u011fer elemanlar\u0131n kayd\u0131r\u0131lmas\u0131 gerekir.<\/p>\n
Temel \u0130\u015flemler:<\/b><\/p>\n\n- Olu\u015fturma:<\/b><\/li>\n
\nmy_list = [1, 2, 3, 4, 5]\nempty_list = []\n <\/pre>\n<\/code><\/p>\n
- Eri\u015fim:<\/b><\/li>\n
\nprint(my_list[0]) # \u00c7\u0131kt\u0131: 1 (O(1))\nprint(my_list[len(my_list) - 1]) # \u00c7\u0131kt\u0131: 5\n <\/pre>\n<\/code><\/p>\n
- Eleman Ekleme:<\/b><\/li>\n
\nmy_list.append(6) # Listenin sonuna ekler (amortized O(1))\nmy_list.insert(0, 0) # Belirtilen indekse ekler (O(n))\nprint(my_list) # \u00c7\u0131kt\u0131: [0, 1, 2, 3, 4, 5, 6]\n <\/pre>\n<\/code><\/p>\n
- Eleman Silme:<\/b><\/li>\n
\nmy_list.pop() # Sondan eleman siler ve d\u00f6nd\u00fcr\u00fcr (O(1))\nmy_list.pop(0) # Belirtilen indeksteki eleman\u0131 siler ve d\u00f6nd\u00fcr\u00fcr (O(n))\nmy_list.remove(3) # De\u011feri 3 olan ilk eleman\u0131 siler (O(n))\nprint(my_list) # \u00c7\u0131kt\u0131: [1, 2, 4, 5] (\u00f6rne\u011fe g\u00f6re de\u011fi\u015fir)\n <\/pre>\n<\/code><\/p>\n
- Dilimleme (Slicing):<\/b><\/li>\n
\nsub_list = my_list[1:3] # \u0130ndeks 1'den 3'e kadar (3 dahil de\u011fil)\nprint(sub_list) # \u00c7\u0131kt\u0131: [2, 4] (my_list'in son haline g\u00f6re)\n <\/pre>\n<\/code>\n<\/ul>\n
Python listeleri, bir\u00e7ok senaryoda ilk tercihiniz olmal\u0131d\u0131r. Ancak \u00e7ok s\u0131k ba\u015fa veya ortas\u0131na eleman ekleyip \u00e7\u0131karmak durumunda kal\u0131rsan\u0131z, <\/p>\n
collections.deque<\/pre>\n<\/code> gibi \u00e7ift u\u00e7lu bir kuyruk yap\u0131s\u0131 daha performansl\u0131 olabilir.<\/p>\n
Ba\u011fl\u0131 Listeler (Linked Lists): Dinamik Bellek Y\u00f6netimi Nas\u0131l Sa\u011flan\u0131r?<\/h3>\n
Ba\u011fl\u0131 liste, elemanlar\u0131n bellekte biti\u015fik olarak saklanmad\u0131\u011f\u0131, bunun yerine her bir eleman\u0131n (d\u00fc\u011f\u00fcm\u00fcn) hem kendi verisini hem de bir sonraki eleman\u0131n referans\u0131n\u0131 (pointer\u0131n\u0131) i\u00e7erdi\u011fi bir veri yap\u0131s\u0131d\u0131r. Bu yap\u0131, dizilere g\u00f6re ekleme ve silme i\u015flemlerini <\/p>\n
O(1)<\/pre>\n<\/code> zaman karma\u015f\u0131kl\u0131\u011f\u0131nda yapabilme avantaj\u0131 sunar (d\u00fc\u011f\u00fcm\u00fcn konumunu bildi\u011finiz s\u00fcrece), ancak elemanlara eri\u015fim <\/p>\n
O(n)<\/pre>\n<\/code> zaman al\u0131r \u00e7\u00fcnk\u00fc ba\u015ftan itibaren her d\u00fc\u011f\u00fcm\u00fc tek tek gezmeniz gerekir.<\/p>\n
\nclass Node:\n def __init__(self, data):\n self.data = data\n self.next = None\n\nclass LinkedList:\n def __init__(self):\n self.head = None\n\n def append(self, data):\n new_node = Node(data)\n if not self.head:\n self.head = new_node\n return\n last_node = self.head\n while last_node.next:\n last_node = last_node.next\n last_node.next = new_node\n\n def prepend(self, data):\n new_node = Node(data)\n new_node.next = self.head\n self.head = new_node\n\n def delete_node(self, key):\n current_node = self.head\n if current_node and current_node.data == key:\n self.head = current_node.next\n current_node = None\n return\n\n prev = None\n while current_node and current_node.data != key:\n prev = current_node\n current_node = current_node.next\n\n if not current_node:\n print(f\"'{key}' listede bulunamad\u0131.\")\n return\n\n prev.next = current_node.next\n current_node = None\n\n def print_list(self):\n current = self.head\n while current:\n print(current.data, end=\" -> \")\n current = current.next\n print(\"None\")\n\n# Kullan\u0131m \u00f6rne\u011fi\nmy_linked_list = LinkedList()\nmy_linked_list.append(1)\nmy_linked_list.append(2)\nmy_linked_list.prepend(0)\nmy_linked_list.print_list() # \u00c7\u0131kt\u0131: 0 -> 1 -> 2 -> None\nmy_linked_list.delete_node(1)\nmy_linked_list.print_list() # \u00c7\u0131kt\u0131: 0 -> 2 -> None\n <\/pre>\n<\/code><\/p>\n
Ba\u011fl\u0131 listeler, \u00f6zellikle dinamik bellek gerektiren durumlarda, \u00f6rne\u011fin bir metin d\u00fczenleyicide geri alma\/ileri alma (undo\/redo) i\u015flevleri veya oyunlardaki karakter envanteri gibi senaryolarda tercih edilebilir.<\/p>\n
Y\u0131\u011f\u0131nlar (Stacks) ve Kuyruklar (Queues): LIFO ve FIFO Prensipleriyle \u00c7al\u0131\u015fmak<\/h3>\n
Y\u0131\u011f\u0131nlar (Stacks) ve kuyruklar (Queues), belirli eri\u015fim kurallar\u0131na sahip \u00f6zel t\u00fcrde listelerdir.<\/p>\n
Y\u0131\u011f\u0131nlar (Stacks - LIFO)<\/h4>\n
Y\u0131\u011f\u0131n, Last-In, First-Out (LIFO) prensibiyle \u00e7al\u0131\u015fan bir veri yap\u0131s\u0131d\u0131r. Son eklenen eleman, ilk \u00e7\u0131kar\u0131lan elemand\u0131r. G\u00fcnl\u00fck hayatta \u00fcst \u00fcste y\u0131\u011f\u0131lm\u0131\u015f tabaklar iyi bir \u00f6rnektir. Python'da y\u0131\u011f\u0131nlar\u0131 genellikle bir liste kullanarak implemente ederiz.<\/p>\n
\nstack = []\nstack.append('A') # Ekleme (push)\nstack.append('B')\nstack.append('C')\nprint(stack) # \u00c7\u0131kt\u0131: ['A', 'B', 'C']\nprint(stack.pop()) # \u00c7\u0131kt\u0131: C (\u00c7\u0131karma - pop)\nprint(stack.pop()) # \u00c7\u0131kt\u0131: B\nprint(stack) # \u00c7\u0131kt\u0131: ['A']\n <\/pre>\n<\/code><\/p>\n
Y\u0131\u011f\u0131nlar, fonksiyon \u00e7a\u011fr\u0131 y\u0131\u011f\u0131nlar\u0131nda, parantez denetiminde veya bir web taray\u0131c\u0131s\u0131n\u0131n geri tu\u015fu i\u015flevselli\u011finde kullan\u0131l\u0131r.<\/p>\n
Kuyruklar (Queues - FIFO)<\/h4>\n
Kuyruk, First-In, First-Out (FIFO) prensibiyle \u00e7al\u0131\u015fan bir veri yap\u0131s\u0131d\u0131r. \u0130lk eklenen eleman, ilk \u00e7\u0131kar\u0131lan elemand\u0131r. S\u00fcpermarket kasas\u0131 s\u0131ras\u0131 bunun en iyi \u00f6rne\u011fidir. Python'da kuyruklar\u0131 <\/p>\n
collections.deque<\/pre>\n<\/code> kullanarak verimli bir \u015fekilde implemente edebiliriz.<\/p>\n
\nfrom collections import deque\n\nqueue = deque()\nqueue.append('A') # Ekleme (enqueue)\nqueue.append('B')\nqueue.append('C')\nprint(queue) # \u00c7\u0131kt\u0131: deque(['A', 'B', 'C'])\nprint(queue.popleft()) # \u00c7\u0131kt\u0131: A (\u00c7\u0131karma - dequeue)\nprint(queue) # \u00c7\u0131kt\u0131: deque(['B', 'C'])\n <\/pre>\n<\/code><\/p>\n
Kuyruklar, yazd\u0131rma kuyruklar\u0131nda, i\u015flem zamanlay\u0131c\u0131larda veya a\u011f paketi iletiminde kullan\u0131l\u0131r.<\/p>\n
Hash Tablolar\u0131 (Dictionaries): O(1) Ortalama S\u00fcrede Eri\u015fim Nas\u0131l Elde Edilir?<\/h3>\n
Python'da s\u00f6zl\u00fckler (dictionaries), hash tablolar\u0131n\u0131n bir implementasyonudur. Anahtar-de\u011fer \u00e7iftlerini depolayan bu yap\u0131lar, anahtarlara g\u00f6re elemanlara ortalama <\/p>\n
O(1)<\/pre>\n<\/code> zamanda eri\u015fim, ekleme ve silme imkan\u0131 sunar. Bu inan\u0131lmaz verimlilik, hash fonksiyonlar\u0131 sayesinde sa\u011flan\u0131r; her anahtar, bellekteki benzersiz bir konuma e\u015flenir.<\/p>\n
\nmy_dict = {\"elma\": 1, \"armut\": 2, \"kiraz\": 3}\n\n# Eleman ekleme\/g\u00fcncelleme (O(1) ortalama)\nmy_dict[\"muz\"] = 4\nprint(my_dict) # \u00c7\u0131kt\u0131: {'elma': 1, 'armut': 2, 'kiraz': 3, 'muz': 4}\n\n# Elemana eri\u015fim (O(1) ortalama)\nprint(my_dict[\"elma\"]) # \u00c7\u0131kt\u0131: 1\n\n# Eleman silme (O(1) ortalama)\ndel my_dict[\"armut\"]\nprint(my_dict) # \u00c7\u0131kt\u0131: {'elma': 1, 'kiraz': 3, 'muz': 4}\n\n# Anahtar\u0131n varl\u0131\u011f\u0131n\u0131 kontrol etme (O(1) ortalama)\nif \"kiraz\" in my_dict:\n print(\"Kiraz var.\")\n <\/pre>\n<\/code><\/p>\n
Hash tablolar\u0131, veritaban\u0131 indekslemede, \u00f6nbelleklemede, frekans say\u0131m\u0131nda veya telefon rehberleri gibi anahtar-de\u011fer ili\u015fkisi olan bir\u00e7ok uygulamada temel bir bile\u015fendir. Python'\u0131n yerle\u015fik s\u00f6zl\u00fckleri o kadar iyi optimize edilmi\u015ftir ki, \u00e7o\u011fu durumda \u00f6zel bir hash tablosu implementasyonuna ihtiya\u00e7 duymazs\u0131n\u0131z.<\/p>\n
Algoritma Temelleri: Python ile Etkili \u00c7\u00f6z\u00fcmler Geli\u015ftirmek<\/h2>\n
Veri yap\u0131lar\u0131n\u0131 anlad\u0131ktan sonra, s\u0131ra bu yap\u0131lar \u00fczerinde \u00e7al\u0131\u015fan algoritmalar\u0131 \u00f6\u011frenmeye gelir. Algoritmalar, problemleri \u00e7\u00f6zmek i\u00e7in izledi\u011fimiz tariflerdir. Bir problemi \u00e7\u00f6zmenin birden fazla yolu olabilir, ancak en iyi algoritma, hem do\u011fru sonucu veren hem de bunu en verimli \u015fekilde (en az zaman ve kaynakla) yapan algoritmad\u0131r. Bu b\u00f6l\u00fcmde, s\u0131k\u00e7a kar\u015f\u0131la\u015f\u0131lan baz\u0131 temel algoritma t\u00fcrlerini ve Python'da nas\u0131l uygulanabileceklerini inceleyece\u011fiz.<\/p>\n
S\u0131ralama Algoritmalar\u0131: Verileri D\u00fczenlemenin En Pop\u00fcler Yollar\u0131 Nelerdir?<\/h3>\n
S\u0131ralama algoritmalar\u0131, bir veri koleksiyonundaki elemanlar\u0131 belirli bir d\u00fczene (artana veya azalana) g\u00f6re yerle\u015ftirmek i\u00e7in kullan\u0131l\u0131r. Bu algoritmalar, veri analizi, veritabanlar\u0131 ve arama i\u015flemleri gibi bir\u00e7ok alanda temel bir rol oynar. Python'\u0131n kendi <\/p>\n
sort()<\/pre>\n<\/code> metodu (listeler i\u00e7in) ve <\/p>\n
sorted()<\/pre>\n<\/code> fonksiyonu (herhangi bir yinelenebilir nesne i\u00e7in) TimSort ad\u0131 verilen hibrit bir algoritma kullan\u0131r ve \u00e7o\u011fu durumda olduk\u00e7a verimlidir (<\/p>\n
O(n log n)<\/pre>\n<\/code> ortalama ve en k\u00f6t\u00fc durum). Ancak, i\u00e7 \u00e7al\u0131\u015fma mekanizmalar\u0131n\u0131 anlamak i\u00e7in baz\u0131 temel s\u0131ralama algoritmalar\u0131na g\u00f6z atal\u0131m.<\/p>\n
Kabarc\u0131k S\u0131ralamas\u0131 (Bubble Sort)<\/h4>\n
Basit ama verimsiz bir algoritmad\u0131r (<\/p>\n
O(n^2)<\/pre>\n<\/code>). Tekrarlayan ge\u00e7i\u015flerle yan yana duran elemanlar\u0131 kar\u015f\u0131la\u015ft\u0131r\u0131r ve e\u011fer yanl\u0131\u015f s\u0131radalarsa yerlerini de\u011fi\u015ftirir.<\/p>\n
\ndef bubble_sort(arr):\n n = len(arr)\n for i in range(n):\n swapped = False\n for j in range(0, n - i - 1):\n if arr[j] > arr[j + 1]:\n arr[j], arr[j + 1] = arr[j + 1], arr[j] # Takas\n swapped = True\n if not swapped: # E\u011fer bu ge\u00e7i\u015fte hi\u00e7bir takas yap\u0131lmad\u0131ysa, dizi s\u0131ralanm\u0131\u015ft\u0131r.\n break\n return arr\n\n# Kullan\u0131m \u00f6rne\u011fi\nmy_arr = [64, 34, 25, 12, 22, 11, 90]\nprint(f\"Kabarc\u0131k S\u0131ralamas\u0131: {bubble_sort(list(my_arr))}\")\n <\/pre>\n<\/code><\/p>\n
Kabarc\u0131k s\u0131ralamas\u0131, k\u00fc\u00e7\u00fck diziler i\u00e7in anla\u015f\u0131lmas\u0131 kolayd\u0131r ancak b\u00fcy\u00fck veri setlerinde asla kullan\u0131lmamal\u0131d\u0131r.<\/p>\n
Birle\u015ftirme S\u0131ralamas\u0131 (Merge Sort)<\/h4>\n
Daha verimli, Divide and Conquer (B\u00f6l ve Y\u00f6net) prensibine dayal\u0131 bir s\u0131ralama algoritmas\u0131d\u0131r (<\/p>\n
O(n log n)<\/pre>\n<\/code>). Diziyi s\u00fcrekli ikiye b\u00f6ler, her par\u00e7ay\u0131 ayr\u0131 ayr\u0131 s\u0131ralar ve sonra s\u0131ralanm\u0131\u015f par\u00e7alar\u0131 birle\u015ftirir.<\/p>\n
\ndef merge_sort(arr):\n if len(arr) > 1:\n mid = len(arr) \/\/ 2\n left_half = arr[:mid]\n right_half = arr[mid:]\n\n merge_sort(left_half)\n merge_sort(right_half)\n\n i = j = k = 0\n\n while i < len(left_half) and j < len(right_half):\n if left_half[i] < right_half[j]:\n arr[k] = left_half[i]\n i += 1\n else:\n arr[k] = right_half[j]\n j += 1\n k += 1\n\n while i < len(left_half):\n arr[k] = left_half[i]\n i += 1\n k += 1\n\n while j < len(right_half):\n arr[k] = right_half[j]\n j += 1\n k += 1\n return arr\n\n# Kullan\u0131m \u00f6rne\u011fi\nprint(f\"Birle\u015ftirme S\u0131ralamas\u0131: {merge_sort(list(my_arr))}\")\n <\/pre>\n<\/code><\/p>\n
Birle\u015ftirme s\u0131ralamas\u0131, b\u00fcy\u00fck veri setleri i\u00e7in stabil ve verimlidir, ancak ek bellek gerektirir.<\/p>\n
Arama Algoritmalar\u0131: Hedefi H\u0131zl\u0131ca Bulmak \u0130\u00e7in Hangi Y\u00f6ntemleri Kullanmal\u0131y\u0131z?<\/h3>\n
Arama algoritmalar\u0131, bir veri koleksiyonunda belirli bir \u00f6\u011fenin varl\u0131\u011f\u0131n\u0131 kontrol etmek veya konumunu bulmak i\u00e7in kullan\u0131l\u0131r. Veri yap\u0131lar\u0131 ile birlikte, bu algoritmalar herhangi bir veri odakl\u0131 uygulaman\u0131n temelini olu\u015fturur.<\/p>\n
Do\u011frusal Arama (Linear Search)<\/h4>\n
En basit arama algoritmas\u0131d\u0131r (<\/p>\n
O(n)<\/pre>\n<\/code>). Dizideki her eleman\u0131 tek tek kontrol ederek hedefi bulmaya \u00e7al\u0131\u015f\u0131r. Dizi s\u0131ral\u0131 olsun veya olmas\u0131n kullan\u0131labilir.<\/p>\n
\ndef linear_search(arr, target):\n for i in range(len(arr)):\n if arr[i] == target:\n return i # Hedef bulundu, indeksi d\u00f6nd\u00fcr\n return -1 # Hedef bulunamad\u0131\n\n# Kullan\u0131m \u00f6rne\u011fi\nprint(f\"Do\u011frusal Arama (22): {linear_search(my_arr, 22)}\") # \u00c7\u0131kt\u0131: 4 (my_arr s\u0131ralanmam\u0131\u015f hali)\nprint(f\"Do\u011frusal Arama (100): {linear_search(my_arr, 100)}\") # \u00c7\u0131kt\u0131: -1\n <\/pre>\n<\/code><\/p>\n
\u0130kili Arama (Binary Search)<\/h4>\n
Daha verimli bir arama algoritmas\u0131d\u0131r (<\/p>\n
O(log n)<\/pre>\n<\/code>), ancak sadece s\u0131ral\u0131 dizilerde<\/strong> \u00e7al\u0131\u015f\u0131r. Diziyi s\u00fcrekli ikiye b\u00f6lerek hedefin hangi yar\u0131da oldu\u011funu kontrol eder ve arama alan\u0131n\u0131 daralt\u0131r.<\/p>\n\ndef binary_search(arr, target):\n low = 0\n high = len(arr) - 1\n\n while low <= high:\n mid = (low + high) \/\/ 2\n if arr[mid] == target:\n return mid\n elif arr[mid] < target:\n low = mid + 1\n else:\n high = mid - 1\n return -1\n\n# Kullan\u0131m \u00f6rne\u011fi (dizi s\u0131ral\u0131 olmal\u0131!)\nsorted_arr = sorted(my_arr) # my_arr'\u0131 s\u0131ralayal\u0131m\nprint(f\"\u0130kili Arama (22, s\u0131ral\u0131): {binary_search(sorted_arr, 22)}\")\nprint(f\"\u0130kili Arama (100, s\u0131ral\u0131): {binary_search(sorted_arr, 100)}\")\n <\/pre>\n<\/code><\/p>\n
\u0130kili arama, b\u00fcy\u00fck s\u0131ral\u0131 veri setlerinde h\u0131zl\u0131 arama gerekti\u011finde vazge\u00e7ilmezdir. Telefon rehberleri, s\u00f6zl\u00fck uygulamalar\u0131 gibi yerlerde yayg\u0131n olarak kullan\u0131l\u0131r.<\/p>\n
Tekrarlamal\u0131 (Recursion) ve \u0130teratif (Iteration) Yakla\u015f\u0131mlar: Ne Zaman Hangisi Tercih Edilmeli?<\/h3>\n
Bir problemi \u00e7\u00f6zmek i\u00e7in iki ana programlama yakla\u015f\u0131m\u0131 vard\u0131r: tekrarlamal\u0131 (recursion) ve iteratif (iteration). Her ikisinin de avantajlar\u0131 ve dezavantajlar\u0131 bulunur.<\/p>\n
Tekrarlamal\u0131 (Recursion)<\/h4>\n
Bir fonksiyonun kendi kendini \u00e7a\u011f\u0131rmas\u0131d\u0131r. Problemi daha k\u00fc\u00e7\u00fck, benzer alt problemlere b\u00f6lerek \u00e7\u00f6zer. Temel bir durum (base case) tan\u0131mlanana kadar bu i\u015flem devam eder. Okunabilirli\u011fi y\u00fcksek, \u00f6zellikle a\u011fa\u00e7 ve grafik gibi do\u011fal olarak \u00f6zyinelemeli veri yap\u0131lar\u0131nda zarif \u00e7\u00f6z\u00fcmler sunar.<\/p>\n
\ndef factorial_recursive(n):\n if n == 0: # Temel durum\n return 1\n else:\n return n * factorial_recursive(n - 1) # \u00d6zyinelemeli \u00e7a\u011fr\u0131\n\nprint(f\"Fakt\u00f6riyel (\u00d6zyinelemeli): {factorial_recursive(5)}\") # \u00c7\u0131kt\u0131: 120\n <\/pre>\n<\/code><\/p>\n
Dezavantajlar\u0131 aras\u0131nda artan bellek kullan\u0131m\u0131 (her fonksiyon \u00e7a\u011fr\u0131s\u0131 y\u0131\u011f\u0131nda yer kaplar) ve potansiyel y\u0131\u011f\u0131n ta\u015fmas\u0131 (stack overflow) riski bulunur.<\/p>\n
\u0130teratif (Iteration)<\/h4>\n
for<\/pre>\n<\/code> veya <\/p>\n
while<\/pre>\n<\/code> d\u00f6ng\u00fcleri kullanarak bir i\u015flemi tekrarlamakt\u0131r. Genellikle daha az bellek kullan\u0131r ve y\u0131\u011f\u0131n ta\u015fmas\u0131 riski ta\u015f\u0131maz.<\/p>\n
\ndef factorial_iterative(n):\n res = 1\n for i in range(1, n + 1):\n res *= i\n return res\n\nprint(f\"Fakt\u00f6riyel (\u0130teratif): {factorial_iterative(5)}\") # \u00c7\u0131kt\u0131: 120\n <\/pre>\n<\/code><\/p>\n
Se\u00e7im, problemin do\u011fas\u0131na ve ki\u015fisel tercihe ba\u011fl\u0131d\u0131r. Baz\u0131 problemler \u00f6zyinelemeyle daha do\u011fal ve anla\u015f\u0131l\u0131r bir \u015fekilde ifade edilirken, di\u011ferleri i\u00e7in d\u00f6ng\u00fc tabanl\u0131 bir \u00e7\u00f6z\u00fcm daha verimli veya g\u00fcvenli olabilir. Genellikle, m\u00fcmk\u00fcnse iteratif \u00e7\u00f6z\u00fcmler tercih edilir, ancak \u00f6zyinelemeli \u00e7\u00f6z\u00fcmlerin g\u00fczelli\u011fi ve sadeli\u011fi yads\u0131namaz.<\/p>\n
\n Uzman \u0130pucu: Recursion kullan\u0131rken her zaman bir temel durum (base case) tan\u0131mlad\u0131\u011f\u0131n\u0131zdan emin olun. Aksi takdirde sonsuz d\u00f6ng\u00fcye girer ve \"RecursionError: maximum recursion depth exceeded\" hatas\u0131 al\u0131rs\u0131n\u0131z.\n<\/div>\n\u0130leri D\u00fczey Konular ve Optimizasyon Stratejileri<\/h2>\n
Temel veri yap\u0131lar\u0131 ve algoritmalar\u0131 \u00f6\u011frendikten sonra, s\u0131ra daha karma\u015f\u0131k konulara ve performans\u0131 optimize etme yollar\u0131na gelir. B\u00fcy\u00fck \u00f6l\u00e7ekli sistemler ve rekabet\u00e7i programlama ortamlar\u0131, sadece do\u011fru sonucu veren de\u011fil, ayn\u0131 zamanda bunu en h\u0131zl\u0131 ve en az kaynakla yapan \u00e7\u00f6z\u00fcmleri gerektirir. Bu b\u00f6l\u00fcmde, daha geli\u015fmi\u015f veri yap\u0131lar\u0131n\u0131, algoritmalar\u0131n performans\u0131n\u0131 \u00f6l\u00e7me y\u00f6ntemlerini ve Python'da optimizasyon i\u00e7in baz\u0131 pratik ipu\u00e7lar\u0131n\u0131 ke\u015ffedece\u011fiz.<\/p>\n
A\u011fa\u00e7lar (Trees) ve Graflar (Graphs): Karma\u015f\u0131k \u0130li\u015fkileri Modellemek<\/h3>\n
A\u011fa\u00e7lar ve graflar, elemanlar aras\u0131ndaki karma\u015f\u0131k, do\u011frusal olmayan ili\u015fkileri modellemek i\u00e7in kullan\u0131lan g\u00fc\u00e7l\u00fc veri yap\u0131lar\u0131d\u0131r. \u0130nternet, sosyal a\u011flar, yol a\u011flar\u0131 gibi bir\u00e7ok ger\u00e7ek d\u00fcnya senaryosu bu yap\u0131larla temsil edilebilir.<\/p>\n
A\u011fa\u00e7lar (Trees)<\/h4>\n
A\u011fa\u00e7, hiyerar\u015fik bir yap\u0131y\u0131 temsil eden, k\u00f6k (root) d\u00fc\u011f\u00fcmden ba\u015flayan ve her d\u00fc\u011f\u00fcm\u00fcn s\u0131f\u0131r veya daha fazla \u00e7ocuk d\u00fc\u011f\u00fcme sahip oldu\u011fu do\u011frusal olmayan bir veri yap\u0131s\u0131d\u0131r. En yayg\u0131n t\u00fcrlerinden biri \u0130kili Arama A\u011fac\u0131 (Binary Search Tree - BST) olup, elemanlar\u0131n s\u0131ral\u0131 bir \u015fekilde yerle\u015ftirilmesini sa\u011flar, b\u00f6ylece arama, ekleme ve silme i\u015flemleri ortalama <\/p>\n
O(log n)<\/pre>\n<\/code> zamanda ger\u00e7ekle\u015febilir.<\/p>\n
\nclass TreeNode:\n def __init__(self, key):\n self.left = None\n self.right = None\n self.val = key\n\nclass BST:\n def __init__(self):\n self.root = None\n\n def insert(self, root, key):\n if root is None:\n return TreeNode(key)\n else:\n if root.val < key:\n root.right = self.insert(root.right, key)\n else:\n root.left = self.insert(root.left, key)\n return root\n\n def search(self, root, key):\n if root is None or root.val == key:\n return root\n if root.val < key:\n return self.search(root.right, key)\n return self.search(root.left, key)\n\n def inorder_traversal(self, root):\n if root:\n self.inorder_traversal(root.left)\n print(root.val, end=\" \")\n self.inorder_traversal(root.right)\n\n# Kullan\u0131m \u00f6rne\u011fi\nbst = BST()\nroot = None\nkeys = [50, 30, 70, 20, 40, 60, 80]\nfor key in keys:\n root = bst.insert(root, key)\n\nprint(\"Inorder Dola\u015f\u0131m:\", end=\" \")\nbst.inorder_traversal(root) # \u00c7\u0131kt\u0131: 20 30 40 50 60 70 80\nprint(\"\\n50 bulundu mu?\", bst.search(root, 50) is not None)\n <\/pre>\n<\/code><\/p>\n
A\u011fa\u00e7lar, dosya sistemlerinde, veritaban\u0131 indekslemede ve \u00f6ncelik kuyruklar\u0131nda (heap) kullan\u0131l\u0131r.<\/p>\n
Graflar (Graphs)<\/h4>\n
Graf, d\u00fc\u011f\u00fcmler (vertices) ve bu d\u00fc\u011f\u00fcmler aras\u0131ndaki ba\u011flant\u0131lar (edges) k\u00fcmesinden olu\u015fan daha genel bir veri yap\u0131s\u0131d\u0131r. Y\u00f6nl\u00fc veya y\u00f6ns\u00fcz, a\u011f\u0131rl\u0131kl\u0131 veya a\u011f\u0131rl\u0131ks\u0131z olabilirler. Sosyal a\u011flar, haritalar ve ileti\u015fim a\u011flar\u0131 gibi bir\u00e7ok ger\u00e7ek d\u00fcnya a\u011f\u0131, graflar kullan\u0131larak modellenebilir.<\/p>\n
\ngraph = {\n 'A': ['B', 'C'],\n 'B': ['D', 'E'],\n 'C': ['F'],\n 'D': [],\n 'E': ['F'],\n 'F': []\n}\n\n# Geni\u015flik \u00d6ncelikli Arama (BFS)\nfrom collections import deque\n\ndef bfs(graph, start_node):\n visited = set()\n queue = deque([start_node])\n visited.add(start_node)\n\n while queue:\n node = queue.popleft()\n print(node, end=\" \")\n for neighbor in graph[node]:\n if neighbor not in visited:\n visited.add(neighbor)\n queue.append(neighbor)\n\nprint(\"BFS Dola\u015f\u0131m\u0131:\", end=\" \")\nbfs(graph, 'A') # \u00c7\u0131kt\u0131: A B C D E F\nprint()\n\n# Derinlik \u00d6ncelikli Arama (DFS)\ndef dfs(graph, start_node, visited=None):\n if visited is None:\n visited = set()\n visited.add(start_node)\n print(start_node, end=\" \")\n for neighbor in graph[start_node]:\n if neighbor not in visited:\n dfs(graph, neighbor, visited)\n\nprint(\"DFS Dola\u015f\u0131m\u0131:\", end=\" \")\ndfs(graph, 'A') # \u00c7\u0131kt\u0131: A B D E F C\nprint()\n <\/pre>\n<\/code><\/p>\n
Graflar, yol bulma algoritmalar\u0131nda (Dijkstra, A*), sosyal a\u011f analizinde ve a\u011f g\u00fcvenli\u011finde kilit rol oynar.<\/p>\n
Zaman ve Alan Karma\u015f\u0131kl\u0131\u011f\u0131 Analizi (Big O Notasyonu): Kodunuz Ne Kadar Verimli?<\/h3>\n
Bir algoritman\u0131n verimlili\u011fini de\u011ferlendirmek i\u00e7in Zaman Karma\u015f\u0131kl\u0131\u011f\u0131 (Time Complexity) ve Alan Karma\u015f\u0131kl\u0131\u011f\u0131 (Space Complexity) analizi yapar\u0131z. Bu analizler genellikle \"Big O Notasyonu\" kullan\u0131larak ifade edilir. Big O, algoritman\u0131n girdi boyutu b\u00fcy\u00fcd\u00fck\u00e7e \u00e7al\u0131\u015fma s\u00fcresinin veya bellek kullan\u0131m\u0131n\u0131n nas\u0131l artt\u0131\u011f\u0131n\u0131 matematiksel olarak tan\u0131mlayan bir notasyondur.<\/p>\n
\n- \n
O(1)<\/pre>\n<\/code> (Sabit Zaman): Girdi boyutundan ba\u011f\u0131ms\u0131z. \u00d6r: Bir listede belirli bir indekse eri\u015fim.<\/li>\n
- \n
O(log n)<\/pre>\n<\/code> (Logaritmik Zaman): Girdi boyutu artt\u0131k\u00e7a \u00e7al\u0131\u015fma s\u00fcresi yava\u015f\u00e7a artar. \u00d6r: \u0130kili arama.<\/li>\n
- \n
O(n)<\/pre>\n<\/code> (Do\u011frusal Zaman): Girdi boyutuyla do\u011fru orant\u0131l\u0131 olarak artar. \u00d6r: Do\u011frusal arama.<\/li>\n
- \n
O(n log n)<\/pre>\n<\/code> (Do\u011frusal-Logaritmik Zaman): Verimli s\u0131ralama algoritmalar\u0131 (Merge Sort, Quick Sort).<\/li>\n
- \n
O(n^2)<\/pre>\n<\/code> (Karesel Zaman): Girdi boyutunun karesiyle artar. \u00c7ok verimsizdir. \u00d6r: Bubble Sort.<\/li>\n
- \n
O(2^n)<\/pre>\n<\/code> (\u00dcstel Zaman): Girdi boyutundaki k\u00fc\u00e7\u00fck art\u0131\u015flar bile devasa s\u00fcre art\u0131\u015flar\u0131na neden olur. \u00c7ok nadiren kabul edilebilir. \u00d6r: Baz\u0131 brute-force \u00e7\u00f6z\u00fcmler.<\/li>\n<\/ul>\n
Bir algoritmay\u0131 optimize ederken, genellikle zaman karma\u015f\u0131kl\u0131\u011f\u0131n\u0131 d\u00fc\u015f\u00fcrmeye \u00e7al\u0131\u015f\u0131r\u0131z. Ancak bazen zaman kazanmak i\u00e7in daha fazla bellek kullanmam\u0131z gerekebilir (zaman-alan takas\u0131). \u0130yi bir m\u00fchendis, bu takaslar\u0131 do\u011fru bir \u015fekilde de\u011ferlendirebilmelidir.<\/p>\n
Dinamik Programlama: Tekrar Eden Problemleri Ak\u0131ll\u0131ca \u00c7\u00f6zme<\/h3>\n
Dinamik programlama (DP), genellikle \u00fcst \u00fcste binen alt problemlere ve optimal alt yap\u0131ya sahip karma\u015f\u0131k problemleri \u00e7\u00f6zmek i\u00e7in kullan\u0131lan g\u00fc\u00e7l\u00fc bir tekniktir. Temel fikir, ayn\u0131 alt problemleri tekrar tekrar \u00e7\u00f6zmekten ka\u00e7\u0131nmak i\u00e7in ara sonu\u00e7lar\u0131 depolamak ve yeniden kullanmakt\u0131r. Bu, genellikle memoization (\u00fcstten a\u015fa\u011f\u0131) veya tabulation (alttan yukar\u0131) yakla\u015f\u0131mlar\u0131yla yap\u0131l\u0131r.<\/p>\n
\u00d6rnek: Fibonacci Serisi<\/b><\/p>\n\n# \u00d6zyinelemeli (Memoization ile Dinamik Programlama)\ndef fib_dp(n, memo={}):\n if n in memo:\n return memo[n]\n if n <= 2:\n return 1\n memo[n] = fib_dp(n - 1, memo) + fib_dp(n - 2, memo)\n return memo[n]\n\nprint(f\"Fibonacci (DP): {fib_dp(10)}\") # \u00c7\u0131kt\u0131: 55\n\n# \u0130teratif (Tabulation ile Dinamik Programlama)\ndef fib_tab(n):\n if n <= 2:\n return 1\n fib = [0] * (n + 1)\n fib[1] = 1\n fib[2] = 1\n for i in range(3, n + 1):\n fib[i] = fib[i - 1] + fib[i - 2]\n return fib[n]\n\nprint(f\"Fibonacci (Tabulation): {fib_tab(10)}\") # \u00c7\u0131kt\u0131: 55\n <\/pre>\n<\/code><\/p>\n
Dinamik programlama, s\u0131rt \u00e7antas\u0131 problemi, en uzun ortak alt dizi ve yol bulma problemleri gibi bir\u00e7ok karma\u015f\u0131k optimizasyon probleminde hayati \u00f6neme sahiptir.<\/p>\n
Python Performans \u0130pu\u00e7lar\u0131: Kodunuzu Nas\u0131l H\u0131zland\u0131rabilirsiniz?<\/h3>\n
Python genel olarak h\u0131zl\u0131 bir dil olmasa da, do\u011fru tekniklerle ve veri yap\u0131lar\u0131yla performans\u0131 \u00f6nemli \u00f6l\u00e7\u00fcde art\u0131rabilirsiniz.<\/p>\n
\n- Do\u011fru Veri Yap\u0131s\u0131n\u0131 Se\u00e7in:<\/b> Listenin ba\u015f\u0131na s\u0131k\u00e7a ekleme\/silme yap\u0131yorsan\u0131z\n
collections.deque<\/pre>\n<\/code> kullan\u0131n. H\u0131zl\u0131 anahtar-de\u011fer aramalar\u0131 i\u00e7in <\/p>\n
dict<\/pre>\n<\/code> kullan\u0131n.<\/li>\n
- Yerle\u015fik Fonksiyonlar\u0131 ve K\u00fct\u00fcphaneleri Kullan\u0131n:<\/b> Python'\u0131n C ile optimize edilmi\u015f yerle\u015fik fonksiyonlar\u0131 ve k\u00fct\u00fcphaneleri (\u00f6rne\u011fin\n
sum()<\/pre>\n<\/code>, <\/p>\n
min()<\/pre>\n<\/code>, <\/p>\n
max()<\/pre>\n<\/code>, <\/p>\n
sorted()<\/pre>\n<\/code>) genellikle kendi yazd\u0131\u011f\u0131n\u0131z d\u00f6ng\u00fclerden daha h\u0131zl\u0131d\u0131r.<\/li>\n
- List Comprehension Kullan\u0131n:<\/b> D\u00f6ng\u00fc tabanl\u0131 liste olu\u015fturma yerine list comprehension, daha okunakl\u0131 ve genellikle daha h\u0131zl\u0131d\u0131r.<\/li>\n
\n# K\u00f6t\u00fc:\nsquares = []\nfor i in range(10):\n squares.append(i * i)\n\n# \u0130yi:\nsquares_comp = [i * i for i in range(10)]\n <\/pre>\n<\/code><\/p>\n
- JIT Derleyicileri (\u00f6rn. Numba) Kullan\u0131n:<\/b> Yo\u011fun say\u0131sal i\u015flemler i\u00e7eren kod bloklar\u0131 i\u00e7in Numba gibi Just-In-Time (JIT) derleyicileri kullanarak Python kodunuzu neredeyse C h\u0131z\u0131yla \u00e7al\u0131\u015ft\u0131rabilirsiniz.<\/li>\n
\nfrom numba import jit\n\n@jit(nopython=True)\ndef fast_sum(arr):\n total = 0\n for x in arr:\n total += x\n return total\n <\/pre>\n<\/code><\/p>\n
- Bellek Kullan\u0131m\u0131n\u0131 Optimize Edin:<\/b> B\u00fcy\u00fck veri setleriyle \u00e7al\u0131\u015f\u0131rken, jenerat\u00f6rler (generators) gibi lazy evaluation tekniklerini kullanarak bellekteki y\u00fck\u00fc azaltabilirsiniz.<\/li>\n
\n# Jenerat\u00f6r: Bellekte t\u00fcm listeyi tutmaz\ndef fibonacci_generator(limit):\n a, b = 0, 1\n while a < limit:\n yield a\n a, b = b, a + b\n\nfor num in fibonacci_generator(10):\n print(num, end=\" \") # \u00c7\u0131kt\u0131: 0 1 1 2 3 5 8\nprint()\n <\/pre>\n<\/code><\/p>\n
- Profileleme Ara\u00e7lar\u0131 Kullan\u0131n:<\/b> Kodunuzun neresinde zaman harcand\u0131\u011f\u0131n\u0131 anlamak i\u00e7in\n
cProfile<\/pre>\n<\/code> mod\u00fcl\u00fcn\u00fc kullan\u0131n. \"Erken optimizasyon t\u00fcm k\u00f6t\u00fcl\u00fcklerin anas\u0131d\u0131r\" s\u00f6z\u00fcn\u00fc unutmay\u0131n; \u00f6nce profilleyin, sonra optimize edin.<\/li>\n<\/ol>\n
Mobil Uyumluluk Notu (CSS Media Query \u00d6rne\u011fi):<\/b> Her ne kadar bu makale sadece <\/code> i\u00e7eri\u011fini sa\u011flasa da, modern web sayfalar\u0131n\u0131n mobil uyumlu olmas\u0131 esast\u0131r. Yazd\u0131\u011f\u0131n\u0131z kod \u00f6rnekleri veya veri g\u00f6rselle\u015ftirmeleri gibi \u00f6\u011felerin k\u00fc\u00e7\u00fck ekranlarda da d\u00fczg\u00fcn g\u00f6r\u00fcnmesini sa\u011flamak i\u00e7in CSS media query'leri kullanmal\u0131s\u0131n\u0131z. \u00d6rne\u011fin, a\u015fa\u011f\u0131daki CSS kodu, belirli bir ekran geni\u015fli\u011finin alt\u0131nda paragraf yaz\u0131 tipini k\u00fc\u00e7\u00fclt\u00fcr:<\/p>\n