{"id":31607,"date":"2025-10-11T21:41:17","date_gmt":"2025-10-11T18:41:17","guid":{"rendered":"https:\/\/fatihsoysal.com\/blog\/?p=31607"},"modified":"2025-10-11T21:41:17","modified_gmt":"2025-10-11T18:41:17","slug":"yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic","status":"publish","type":"post","link":"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/","title":{"rendered":"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7"},"content":{"rendered":"

<\/p>\n

Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7<\/h2>\n

Yapay zeka (AI) ve makine \u00f6\u011frenimi (ML) alanlar\u0131, g\u00fcn\u00fcm\u00fcz\u00fcn en h\u0131zl\u0131 geli\u015fen ve d\u00f6n\u00fc\u015ft\u00fcr\u00fcc\u00fc teknolojileri aras\u0131nda yer almaktad\u0131r. Bu alanlarda yap\u0131lan \u00e7al\u0131\u015fmalar, b\u00fcy\u00fck veri k\u00fcmelerinin i\u015flenmesini ve karma\u015f\u0131k algoritmalar\u0131n e\u011fitilmesini gerektirir. Geleneksel merkezi i\u015flem birimleri (CPU’lar) bu t\u00fcr yo\u011fun hesaplama g\u00f6revleri i\u00e7in yeterli performans\u0131 sa\u011flayamazken, grafik i\u015flem birimleri (GPU’lar) paralel i\u015fleme yetenekleri sayesinde AI\/ML modellerinin e\u011fitim s\u00fcresini dramatik bir \u015fekilde k\u0131saltmaktad\u0131r. Bu nedenle, AI\/ML geli\u015ftiricileri i\u00e7in g\u00fc\u00e7l\u00fc bir GPU ortam\u0131na sahip olmak vazge\u00e7ilmezdir.<\/p>\n

Yerel donan\u0131m yat\u0131r\u0131m\u0131 yapmak, \u00f6zellikle ba\u015flang\u0131\u00e7 seviyesindeki geli\u015ftiriciler veya b\u00fct\u00e7e k\u0131s\u0131tlamas\u0131 olanlar i\u00e7in maliyetli ve karma\u015f\u0131k olabilir. \u0130\u015fte bu noktada bulut tabanl\u0131 GPU hizmetleri devreye girer. DigitalOcean gibi bulut sa\u011flay\u0131c\u0131lar\u0131, geli\u015ftiricilere h\u0131zl\u0131, esnek ve uygun maliyetli GPU kaynaklar\u0131 sunarak bu engeli ortadan kald\u0131r\u0131r. Bu makalede, DigitalOcean GPU Droplet \u00fczerinde eksiksiz bir AI\/ML geli\u015ftirme ortam\u0131 kurmay\u0131, NVIDIA s\u00fcr\u00fcc\u00fclerini ve CUDA Toolkit’i yap\u0131land\u0131rmay\u0131, Python ortam\u0131n\u0131 haz\u0131rlamay\u0131 ve son olarak Jupyter Labs’\u0131 kurup g\u00fcvenli bir \u015fekilde eri\u015fmeyi ad\u0131m ad\u0131m inceleyece\u011fiz. Bu rehber sayesinde, AI\/ML projelerinizi bulutun g\u00fcc\u00fcyle hayata ge\u00e7irmeye haz\u0131r, tam donan\u0131ml\u0131 bir ortama sahip olacaks\u0131n\u0131z.<\/p>\n

Neden DigitalOcean GPU Droplet’leri AI\/ML \u0130\u00e7in \u0130dealdir?<\/h3>\n

AI\/ML geli\u015ftirme s\u00fcre\u00e7lerinde, model e\u011fitimi ve veri analizi gibi yo\u011fun hesaplama gerektiren g\u00f6revler i\u00e7in GPU’lar kritik \u00f6neme sahiptir. B\u00fcy\u00fck bulut sa\u011flay\u0131c\u0131lar\u0131 (AWS, Google Cloud, Azure) geni\u015f GPU se\u00e7enekleri sunsa da, genellikle karma\u015f\u0131k aray\u00fczleri, fiyatland\u0131rma modelleri ve \u00f6\u011frenme e\u011frileri ba\u015flang\u0131\u00e7 seviyesindeki kullan\u0131c\u0131lar i\u00e7in y\u0131ld\u0131r\u0131c\u0131 olabilir. DigitalOcean ise, kullan\u0131c\u0131 dostu aray\u00fcz\u00fc, \u015feffaf ve \u00f6ng\u00f6r\u00fclebilir fiyatland\u0131rmas\u0131 ve basit droplet (sanal sunucu) olu\u015fturma s\u00fcreci ile \u00f6ne \u00e7\u0131kar.<\/p>\n

DigitalOcean’\u0131n GPU Droplet’leri, \u00f6zellikle h\u0131zl\u0131 prototipleme, k\u00fc\u00e7\u00fck ve orta \u00f6l\u00e7ekli projeler veya bireysel geli\u015ftiriciler i\u00e7in idealdir. A\u015fa\u011f\u0131da DigitalOcean’\u0131n sundu\u011fu baz\u0131 avantajlar listelenmi\u015ftir:<\/p>\n

* Basitlik ve Kullan\u0131m Kolayl\u0131\u011f\u0131:<\/strong> DigitalOcean’\u0131n paneli, droplet olu\u015fturma ve y\u00f6netme s\u00fcre\u00e7lerini olduk\u00e7a basitle\u015ftirir. Karma\u015f\u0131k a\u011f veya g\u00fcvenlik yap\u0131land\u0131rmalar\u0131yla u\u011fra\u015fmak zorunda kalmadan dakikalar i\u00e7inde bir GPU sunucusu ba\u015flatabilirsiniz.
\n* \u00d6ng\u00f6r\u00fclebilir Fiyatland\u0131rma:<\/strong> Di\u011fer bulut sa\u011flay\u0131c\u0131lar\u0131n\u0131n aksine, DigitalOcean’\u0131n fiyatland\u0131rmas\u0131 genellikle daha \u015feffaf ve sabittir. Bu, b\u00fct\u00e7enizi daha kolay y\u00f6netmenizi ve beklenmedik maliyetlerle kar\u015f\u0131la\u015fmaman\u0131z\u0131 sa\u011flar.
\n* H\u0131zl\u0131 Da\u011f\u0131t\u0131m:<\/strong> Birka\u00e7 t\u0131klama ile saniyeler i\u00e7inde yeni bir GPU droplet’i olu\u015fturup SSH \u00fczerinden eri\u015fime haz\u0131r hale getirebilirsiniz.
\n* Geli\u015ftirici Dostu:<\/strong> DigitalOcean, geli\u015ftiricilerin ihtiya\u00e7lar\u0131n\u0131 \u00f6n planda tutan bir platformdur. Kapsaml\u0131 dok\u00fcmantasyonlar\u0131 ve topluluk deste\u011fi, kar\u015f\u0131la\u015f\u0131labilecek sorunlar\u0131n \u00e7\u00f6z\u00fcm\u00fcnde yard\u0131mc\u0131 olur.
\n* Esneklik:<\/strong> \u0130htiya\u00e7lar\u0131n\u0131za g\u00f6re farkl\u0131 GPU t\u00fcrleri ve CPU\/RAM kombinasyonlar\u0131 sunulur. Projenizin gereksinimleri de\u011fi\u015ftik\u00e7e droplet’inizi kolayca y\u00fckseltebilir veya k\u00fc\u00e7\u00fcltebilirsiniz.<\/p>\n

Bu avantajlar, DigitalOcean GPU Droplet’lerini AI\/ML projelerinize h\u0131zl\u0131 ve verimli bir ba\u015flang\u0131\u00e7 yapmak i\u00e7in cazip bir se\u00e7enek haline getirmektedir.<\/p>\n

DigitalOcean GPU Droplet Olu\u015fturma Ad\u0131mlar\u0131<\/h3>\n

AI\/ML geli\u015ftirme ortam\u0131m\u0131z\u0131n temelini olu\u015fturacak GPU Droplet’imizi DigitalOcean \u00fczerinde ad\u0131m ad\u0131m olu\u015ftural\u0131m.<\/p>\n

DigitalOcean Hesab\u0131na Giri\u015f ve Droplet Olu\u015fturma<\/h4>\n

1. DigitalOcean Hesab\u0131 Olu\u015fturma\/Giri\u015f Yapma:<\/strong> E\u011fer bir DigitalOcean hesab\u0131n\u0131z yoksa, \u00f6ncelikle digitalocean.com<\/code> adresinden bir hesap olu\u015fturman\u0131z gerekmektedir. Mevcut bir hesab\u0131n\u0131z varsa, giri\u015f yap\u0131n.
\n2. Droplet Olu\u015fturma:<\/strong> Kontrol panelinde sa\u011f \u00fcst k\u00f6\u015fede bulunan “Create” butonuna t\u0131klay\u0131n ve a\u00e7\u0131lan men\u00fcden “Droplets” se\u00e7ene\u011fini se\u00e7in.<\/p>\n

B\u00f6lge ve GPU Plan\u0131 Se\u00e7imi<\/h4>\n

1. B\u00f6lge Se\u00e7imi:<\/strong> GPU Droplet’lerinin kullan\u0131labilir oldu\u011fu belirli b\u00f6lgeler vard\u0131r (\u00f6rne\u011fin, Amsterdam, Frankfurt, New York, San Francisco). Size co\u011frafi olarak en yak\u0131n veya GPU kaynaklar\u0131n\u0131n bol oldu\u011fu bir b\u00f6lgeyi se\u00e7in. B\u00f6lge se\u00e7imi, gecikmeyi (latency) etkileyebilir.
\n2. Droplet Tipi Se\u00e7imi:<\/strong> “Choose a Droplet type” b\u00f6l\u00fcm\u00fcnde “GPU Optimized” se\u00e7ene\u011fini i\u015faretleyin. Bu, sistemin size GPU destekli planlar\u0131 g\u00f6stermesini sa\u011flayacakt\u0131r.
\n3. GPU Plan\u0131 Se\u00e7imi:<\/strong> DigitalOcean, farkl\u0131 GPU donan\u0131mlar\u0131 (\u00f6rne\u011fin, NVIDIA A100, NVIDIA L4) ile \u00e7e\u015fitli planlar sunar. Projenizin ihtiya\u00e7lar\u0131na ve b\u00fct\u00e7enize uygun olan\u0131 se\u00e7in. Daha b\u00fcy\u00fck modeller e\u011fitmek veya daha h\u0131zl\u0131 hesaplama yapmak i\u00e7in daha fazla VRAM’e (GPU belle\u011fi) ve \u00e7ekirdek say\u0131s\u0131na sahip planlar\u0131 tercih edebilirsiniz. \u0130lk ba\u015flayanlar i\u00e7in orta seviye bir plan genellikle yeterli olacakt\u0131r.<\/p>\n

\u0130\u015fletim Sistemi ve Kimlik Do\u011frulama<\/h4>\n

1. \u0130\u015fletim Sistemi Se\u00e7imi:<\/strong> “Choose an image” b\u00f6l\u00fcm\u00fcnde, AI\/ML geli\u015ftirme i\u00e7in en yayg\u0131n ve iyi desteklenen i\u015fletim sistemi olan Ubuntu’nun uzun d\u00f6nem destek (LTS) s\u00fcr\u00fcm\u00fcn\u00fc se\u00e7in (\u00f6rne\u011fin, Ubuntu 22.04 LTS veya 20.04 LTS). Baz\u0131 durumlarda DigitalOcean, NVIDIA s\u00fcr\u00fcc\u00fcleri ve CUDA Toolkit \u00f6nceden y\u00fcklenmi\u015f \u00f6zel g\u00f6r\u00fcnt\u00fcler sunabilir; ancak bu makalede manuel kurulumu ele alaca\u011f\u0131z.
\n2. Kimlik Do\u011frulama (SSH Keys):<\/strong> “Authentication” b\u00f6l\u00fcm\u00fcnde “SSH keys” se\u00e7ene\u011fini i\u015faretlemeniz \u015fiddetle tavsiye edilir. G\u00fcvenlik ve kullan\u0131m kolayl\u0131\u011f\u0131 a\u00e7\u0131s\u0131ndan SSH anahtarlar\u0131 parola tabanl\u0131 eri\u015fimden \u00e7ok daha iyidir.
\n * Yeni SSH Anahtar\u0131 Olu\u015fturma:<\/strong> E\u011fer daha \u00f6nce bir SSH anahtar\u0131n\u0131z yoksa, terminalinizde (Linux\/macOS) veya Git Bash (Windows) \u00fczerinde ssh-keygen -t rsa -b 4096<\/code> komutunu kullanarak bir anahtar olu\u015fturabilirsiniz. Olu\u015fturulan anahtar\u0131n genel (public) k\u0131sm\u0131n\u0131 (~\/.ssh\/id_rsa.pub<\/code> dosyas\u0131n\u0131n i\u00e7eri\u011fi) kopyalay\u0131p DigitalOcean’a ekleyin.
\n * Mevcut Anahtar\u0131 Se\u00e7me:<\/strong> Daha \u00f6nce ekledi\u011finiz bir SSH anahtar\u0131n\u0131z varsa, listeden onu se\u00e7in.
\n <\/em>\u00d6nemli:* SSH anahtar\u0131n\u0131z\u0131n \u00f6zel (private) k\u0131sm\u0131n\u0131 (id_rsa<\/code>) kimseyle payla\u015fmay\u0131n ve g\u00fcvenli bir yerde saklay\u0131n.<\/p>\n

Ek Ayarlar ve Droplet Olu\u015fturma<\/h4>\n

1. Depolama (Block Storage):<\/strong> “Add block storage” b\u00f6l\u00fcm\u00fcnde, b\u00fcy\u00fck veri setleri veya model checkpoint’leri i\u00e7in ek depolama alan\u0131 ekleyebilirsiniz. Bu iste\u011fe ba\u011fl\u0131d\u0131r ancak b\u00fcy\u00fck projeler i\u00e7in \u00f6nerilir.
\n2. Hostname ve Etiketler:<\/strong> Droplet’inize kolayca tan\u0131nabilir bir isim verin (\u00f6rne\u011fin, ai-gpu-droplet<\/code>). \u0130ste\u011fe ba\u011fl\u0131 olarak etiketler (tags) ekleyerek droplet’lerinizi d\u00fczenleyebilirsiniz.
\n3. Proje Se\u00e7imi:<\/strong> Droplet’i hangi projenize ba\u011flayaca\u011f\u0131n\u0131z\u0131 se\u00e7in.
\n4. Olu\u015fturma:<\/strong> T\u00fcm ayarlar\u0131 g\u00f6zden ge\u00e7irdikten sonra “Create Droplet” butonuna t\u0131klay\u0131n. Droplet’iniz birka\u00e7 saniye i\u00e7inde olu\u015fturulacak ve size genel IP adresi atanacakt\u0131r. Bu IP adresini bir yere not al\u0131n, \u00e7\u00fcnk\u00fc sunucunuza ba\u011flanmak i\u00e7in kullanacaks\u0131n\u0131z.<\/p>\n

Art\u0131k GPU Droplet’iniz haz\u0131r. Bir sonraki ad\u0131m, bu droplet’e SSH \u00fczerinden ba\u011flanmak ve temel sunucu yap\u0131land\u0131rmalar\u0131n\u0131 ger\u00e7ekle\u015ftirmektir.<\/p>\n

\u0130lk Sunucu Yap\u0131land\u0131rmas\u0131 ve G\u00fcvenlik<\/h3>\n

Droplet’imiz haz\u0131r oldu\u011funa g\u00f6re, SSH \u00fczerinden ba\u011flanarak temel sistem g\u00fcncellemelerini yapal\u0131m ve g\u00fcvenlik ayarlar\u0131n\u0131 yap\u0131land\u0131ral\u0131m.<\/p>\n

SSH ile Droplet’e Ba\u011flanma<\/h4>\n

Terminalinizi (veya Git Bash’i) a\u00e7\u0131n ve a\u015fa\u011f\u0131daki komutu kullanarak droplet’inize ba\u011flan\u0131n:<\/p>\n

ssh root@your_droplet_ip_address<\/code><\/pre>\n
your_droplet_ip_address<\/code> yerine DigitalOcean panelinizde g\u00f6rd\u00fc\u011f\u00fcn\u00fcz droplet’inizin IP adresini yaz\u0131n. E\u011fer ilk kez ba\u011flan\u0131yorsan\u0131z, sunucunun parmak izini onaylaman\u0131z istenebilir; yes<\/code> yaz\u0131p Enter tu\u015funa bas\u0131n.<\/p>\n
Sistem G\u00fcncelleme<\/h4>\n
Ba\u011fland\u0131ktan sonra, i\u015fletim sisteminin paket listesini g\u00fcncelleyin ve y\u00fckl\u00fc paketleri en son s\u00fcr\u00fcmlerine y\u00fckseltin:<\/p>\n
sudo apt update\nsudo apt upgrade -y<\/code><\/pre>\n
Bu i\u015flem, sistemin kararl\u0131l\u0131\u011f\u0131n\u0131 ve g\u00fcvenli\u011fini sa\u011flamak i\u00e7in \u00f6nemlidir.<\/p>\n
Yeni Bir Kullan\u0131c\u0131 Olu\u015fturma ve Sudo Yetkisi Verme<\/h4>\n
root<\/code> kullan\u0131c\u0131s\u0131 ile do\u011frudan \u00e7al\u0131\u015fmak g\u00fcvenlik a\u00e7\u0131s\u0131ndan \u00f6nerilmez. Yeni bir kullan\u0131c\u0131 olu\u015ftural\u0131m ve ona sudo<\/code> yetkisi verelim:<\/p>\n
sudo adduser yourusername<\/code><\/pre>\n
yourusername<\/code> yerine kullanmak istedi\u011finiz kullan\u0131c\u0131 ad\u0131n\u0131 yaz\u0131n. Parola belirlemeniz ve di\u011fer bilgileri girmeniz istenecektir (\u00e7o\u011fu durumda di\u011fer bilgileri bo\u015f b\u0131rakabilirsiniz).<\/p>\n
Yeni kullan\u0131c\u0131ya sudo<\/code> yetkisi vermek i\u00e7in onu sudo<\/code> grubuna ekleyin:<\/p>\n
sudo usermod -aG sudo yourusername<\/code><\/pre>\n
Art\u0131k yourusername<\/code> ile giri\u015f yapabilir ve sudo<\/code> komutunu kullanarak y\u00f6netici ayr\u0131cal\u0131klar\u0131yla komutlar\u0131 \u00e7al\u0131\u015ft\u0131rabilirsiniz. G\u00fcvenlik i\u00e7in, mevcut SSH oturumunuzu kapat\u0131p yeni kullan\u0131c\u0131 ile ba\u011flanman\u0131z \u00f6nerilir:<\/p>\n
exit # root oturumundan \u00e7\u0131k\u0131\u015f\nssh yourusername@your_droplet_ip_address<\/code><\/pre>\n
E\u011fer yeni kullan\u0131c\u0131ya da SSH anahtar\u0131n\u0131zla ba\u011flanmak istiyorsan\u0131z, root<\/code> kullan\u0131c\u0131s\u0131 olarak ba\u011flan\u0131p SSH anahtar\u0131n\u0131z\u0131 yeni kullan\u0131c\u0131n\u0131n .ssh\/authorized_keys<\/code> dosyas\u0131na kopyalaman\u0131z gerekebilir. Ancak DigitalOcean’da genellikle anahtar, olu\u015fturulan t\u00fcm kullan\u0131c\u0131lar i\u00e7in otomatik olarak ayarlan\u0131r. E\u011fer sorun ya\u015farsan\u0131z, root<\/code> olarak ba\u011flan\u0131p sudo rsync --archive --chown=yourusername:yourusername ~\/.ssh \/home\/yourusername<\/code> komutuyla anahtarlar\u0131 kopyalayabilirsiniz.<\/p>\n
Temel G\u00fcvenlik Duvar\u0131 (UFW) Yap\u0131land\u0131rmas\u0131<\/h4>\n
G\u00fcvenlik duvar\u0131 (firewall), sunucunuza yetkisiz eri\u015fimi engellemek i\u00e7in kritik bir bile\u015fendir. Ubuntu’da UFW (Uncomplicated Firewall) kullan\u0131m\u0131 olduk\u00e7a kolayd\u0131r.<\/p>\n
1.  Varsay\u0131lan Kurallar\u0131 Ayarlama:<\/strong> Gelen t\u00fcm ba\u011flant\u0131lar\u0131 reddedip giden t\u00fcm ba\u011flant\u0131lara izin verin:<\/p>\n
sudo ufw default deny incoming\n    sudo ufw default allow outgoing<\/code><\/pre>\n
2.  SSH Ba\u011flant\u0131s\u0131na \u0130zin Verme:<\/strong> Kendi SSH ba\u011flant\u0131n\u0131z\u0131 kesmemek i\u00e7in SSH portuna (varsay\u0131lan 22) izin verin:<\/p>\n
sudo ufw allow ssh<\/code><\/pre>\n
3.  G\u00fcvenlik Duvar\u0131n\u0131 Etkinle\u015ftirme:<\/strong> Kurallar\u0131 uygulay\u0131n ve g\u00fcvenlik duvar\u0131n\u0131 etkinle\u015ftirin:<\/p>\n
sudo ufw enable<\/code><\/pre>\n
    Etkinle\u015ftirme s\u0131ras\u0131nda bir uyar\u0131 mesaj\u0131 alabilirsiniz; y<\/code> yaz\u0131p Enter tu\u015funa basarak onaylay\u0131n.<\/p>\n
4.  Durumu Kontrol Etme:<\/strong> G\u00fcvenlik duvar\u0131n\u0131n durumunu kontrol edin:<\/p>\n
sudo ufw status<\/code><\/pre>\n
    \u00c7\u0131kt\u0131da Status: active<\/code> ve SSH i\u00e7in bir kural g\u00f6rmelisiniz.<\/p>\n
Bu ad\u0131mlarla sunucunuzun temel yap\u0131land\u0131rmas\u0131n\u0131 ve g\u00fcvenlik \u00f6nlemlerini tamamlad\u0131n\u0131z. Art\u0131k GPU’yu AI\/ML g\u00f6revleri i\u00e7in haz\u0131r hale getirme zaman\u0131.<\/p>\n
NVIDIA S\u00fcr\u00fcc\u00fcleri ve CUDA Toolkit Kurulumu<\/h3>\n
AI\/ML k\u00fct\u00fcphanelerinin GPU’nuzu kullanabilmesi i\u00e7in NVIDIA s\u00fcr\u00fcc\u00fclerinin ve CUDA Toolkit’in do\u011fru bir \u015fekilde kurulmas\u0131 \u015fartt\u0131r.<\/p>\n
NVIDIA S\u00fcr\u00fcc\u00fclerinin Kurulumu<\/h4>\n
1.  Mevcut S\u00fcr\u00fcc\u00fcleri Kontrol Etme:<\/strong> Droplet’inizde zaten NVIDIA s\u00fcr\u00fcc\u00fcleri kurulu olup olmad\u0131\u011f\u0131n\u0131 kontrol etmek i\u00e7in a\u015fa\u011f\u0131daki komutu \u00e7al\u0131\u015ft\u0131r\u0131n:<\/p>\n
nvidia-smi<\/code><\/pre>\n
    E\u011fer komut bulunamazsa veya bir hata mesaj\u0131 al\u0131rsan\u0131z, s\u00fcr\u00fcc\u00fclerin kurulu olmad\u0131\u011f\u0131n\u0131 anlars\u0131n\u0131z. E\u011fer kurulu ise, GPU modelinizi ve s\u00fcr\u00fcc\u00fc s\u00fcr\u00fcm\u00fcn\u00fc g\u00f6steren bir tablo g\u00f6receksiniz. Bu durumda 2. ad\u0131m\u0131 atlayabilirsiniz.<\/p>\n
2.  NVIDIA S\u00fcr\u00fcc\u00fclerini Kurma (Kurulu De\u011filse):<\/strong>
\n    *   NVIDIA Depolar\u0131n\u0131 Ekleme:<\/strong> Ubuntu sisteminize NVIDIA’n\u0131n resmi depolar\u0131n\u0131 ekleyerek en g\u00fcncel s\u00fcr\u00fcc\u00fclere eri\u015febilirsiniz.<\/p>\n
sudo apt update\n        sudo apt install software-properties-common -y\n        sudo add-apt-repository ppa:graphics-drivers\/ppa -y\n        sudo apt update<\/code><\/pre>\n
    *   \u00d6nerilen S\u00fcr\u00fcc\u00fcy\u00fc Y\u00fckleme:<\/strong> ubuntu-drivers<\/code> arac\u0131, sisteminiz i\u00e7in \u00f6nerilen s\u00fcr\u00fcc\u00fcy\u00fc bulman\u0131za yard\u0131mc\u0131 olabilir.<\/p>\n
ubuntu-drivers devices<\/code><\/pre>\n
        Bu komut, sisteminizdeki GPU’lar i\u00e7in \u00f6nerilen s\u00fcr\u00fcc\u00fcleri listeleyecektir (\u00f6rne\u011fin, nvidia-driver-535<\/code>). \u00d6nerilen s\u00fcr\u00fcc\u00fcy\u00fc y\u00fcklemek i\u00e7in:<\/p>\n
sudo apt install nvidia-driver-535 -y # \u00d6rnek olarak 535, sizin i\u00e7in farkl\u0131 olabilir<\/code><\/pre>\n
        Veya do\u011frudan \u00f6nerilen s\u00fcr\u00fcc\u00fcy\u00fc y\u00fcklemek i\u00e7in:<\/p>\n
sudo ubuntu-drivers install<\/code><\/pre>\n
    *   Sistemi Yeniden Ba\u015flatma:<\/strong> S\u00fcr\u00fcc\u00fclerin d\u00fczg\u00fcn bir \u015fekilde etkinle\u015fmesi i\u00e7in sunucuyu yeniden ba\u015flatman\u0131z gerekmektedir:<\/p>\n
sudo reboot<\/code><\/pre>\n
    *   Kurulumu Do\u011frulama:<\/strong> Yeniden ba\u015flatman\u0131n ard\u0131ndan SSH ile tekrar ba\u011flan\u0131n ve nvidia-smi<\/code> komutunu tekrar \u00e7al\u0131\u015ft\u0131r\u0131n. Art\u0131k GPU’nuzun bilgilerini ve y\u00fckl\u00fc s\u00fcr\u00fcc\u00fc s\u00fcr\u00fcm\u00fcn\u00fc g\u00f6rmelisiniz.<\/p>\n
CUDA Toolkit Kurulumu<\/h4>\n
CUDA Toolkit, NVIDIA GPU’lar\u0131nda paralel hesaplama yapmay\u0131 sa\u011flayan bir geli\u015ftirme ortam\u0131d\u0131r. AI\/ML k\u00fct\u00fcphaneleri (PyTorch, TensorFlow) GPU’yu kullanmak i\u00e7in CUDA’ya ihtiya\u00e7 duyar.<\/p>\n
1.  CUDA S\u00fcr\u00fcm\u00fcn\u00fc Belirleme:<\/strong> nvidia-smi<\/code> \u00e7\u0131kt\u0131s\u0131nda “CUDA Version” b\u00f6l\u00fcm\u00fcnde g\u00f6r\u00fcnen s\u00fcr\u00fcm, y\u00fckl\u00fc s\u00fcr\u00fcc\u00fcn\u00fcz\u00fcn destekledi\u011fi en y\u00fcksek CUDA s\u00fcr\u00fcm\u00fcn\u00fc g\u00f6sterir (Runtime CUDA). Y\u00fckleyece\u011finiz CUDA Toolkit s\u00fcr\u00fcm\u00fc, bu s\u00fcr\u00fcmle uyumlu olmal\u0131d\u0131r. \u00d6rne\u011fin, nvidia-smi<\/code> \u00e7\u0131kt\u0131s\u0131nda CUDA 12.2 yaz\u0131yorsa, CUDA Toolkit 12.2 veya daha d\u00fc\u015f\u00fck bir s\u00fcr\u00fcm y\u00fckleyebilirsiniz.<\/p>\n
2.  NVIDIA Developer Sitesinden CUDA Toolkit \u0130ndirme:<\/strong>
\n    *   Taray\u0131c\u0131n\u0131zda developer.nvidia.com\/cuda-downloads<\/code> adresine gidin.
\n    *   \u0130\u015fletim sisteminizi (Linux), mimarinizi (x86_64), da\u011f\u0131t\u0131m\u0131n\u0131z\u0131 (Ubuntu) ve s\u00fcr\u00fcm\u00fcn\u00fcz\u00fc (22.04 veya 20.04) se\u00e7in.
\n    *   “Installer Type” olarak genellikle deb (network)<\/code> veya deb (local)<\/code> se\u00e7ene\u011fi tercih edilir. deb (local)<\/code> daha b\u00fcy\u00fck bir dosya indirir ancak kurulum s\u0131ras\u0131nda internet ba\u011flant\u0131s\u0131 gerektirmez. Bu makalede deb (local)<\/code> \u00fczerinden ilerleyece\u011fiz.
\n    *   Size verilen komutlar\u0131 kopyalay\u0131n ve droplet’inizde \u00e7al\u0131\u015ft\u0131r\u0131n. \u00d6rnek komutlar (CUDA 12.2 i\u00e7in):<\/p>\n
wget https:\/\/developer.download.nvidia.com\/compute\/cuda\/12.2.2\/local_installers\/cuda-repo-ubuntu2204-12-2-local_12.2.2-1_amd64.deb\n        sudo dpkg -i cuda-repo-ubuntu2204-12-2-local_12.2.2-1_amd64.deb\n        sudo cp \/var\/cuda-repo-ubuntu2204-12-2-local\/cuda-*-keyring.gpg \/usr\/share\/keyrings\/\n        sudo apt update\n        sudo apt install cuda -y<\/code><\/pre>\n
        Not:<\/em> Yukar\u0131daki komutlar CUDA 12.2 i\u00e7in bir \u00f6rnektir. NVIDIA web sitesindeki en g\u00fcncel komutlar\u0131 kullanman\u0131z \u00f6nemlidir.<\/p>\n
3.  Ortam De\u011fi\u015fkenlerini Ayarlama:<\/strong> CUDA’n\u0131n sisteminiz taraf\u0131ndan tan\u0131nmas\u0131 i\u00e7in ortam de\u011fi\u015fkenlerini ayarlaman\u0131z gerekir. .bashrc<\/code> dosyan\u0131z\u0131 d\u00fczenleyerek bu de\u011fi\u015fkenleri kal\u0131c\u0131 hale getirin:<\/p>\n
nano ~\/.bashrc<\/code><\/pre>\n
    Dosyan\u0131n sonuna a\u015fa\u011f\u0131daki sat\u0131rlar\u0131 ekleyin (CUDA s\u00fcr\u00fcm\u00fcn\u00fcze g\u00f6re yolu g\u00fcncelleyin, \u00f6rne\u011fin cuda-12.2<\/code>):<\/p>\n
export PATH=\/usr\/local\/cuda-12.2\/bin${PATH:+:${PATH}}\n    export LD_LIBRARY_PATH=\/usr\/local\/cuda-12.2\/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}<\/code><\/pre>\n
    Ctrl+X, Y, Enter tu\u015flar\u0131na basarak kaydedin ve \u00e7\u0131k\u0131n. Ard\u0131ndan .bashrc<\/code> dosyas\u0131n\u0131 etkinle\u015ftirin:<\/p>\n
source ~\/.bashrc<\/code><\/pre>\n
4.  Kurulumu Do\u011frulama:<\/strong> CUDA Toolkit’in do\u011fru bir \u015fekilde kurulup kurulmad\u0131\u011f\u0131n\u0131 kontrol edin:<\/p>\n
nvcc --version<\/code><\/pre>\n
    Bu komut, nvcc<\/code> (CUDA C++ derleyicisi) s\u00fcr\u00fcm\u00fcn\u00fc g\u00f6stermelidir. E\u011fer \u00e7\u0131kt\u0131da CUDA Toolkit s\u00fcr\u00fcm\u00fcn\u00fc g\u00f6r\u00fcyorsan\u0131z, kurulum ba\u015far\u0131l\u0131 demektir.<\/p>\n
Tebrikler! GPU’nuz art\u0131k AI\/ML k\u00fct\u00fcphaneleri taraf\u0131ndan kullan\u0131lmaya haz\u0131r. \u015eimdi Python ortam\u0131m\u0131z\u0131 haz\u0131rlayabiliriz.<\/p>\n
Python Ortam\u0131n\u0131n Haz\u0131rlanmas\u0131 (Miniconda)<\/h3>\n
AI\/ML projelerinde farkl\u0131 k\u00fct\u00fcphane s\u00fcr\u00fcmleri ve ba\u011f\u0131ml\u0131l\u0131klar\u0131 s\u0131k\u00e7a kullan\u0131l\u0131r. Bu karma\u015fay\u0131 y\u00f6netmek ve projelerinizi izole tutmak i\u00e7in Anaconda veya Miniconda gibi ortam y\u00f6neticileri vazge\u00e7ilmezdir. Miniconda, Anaconda’n\u0131n hafif bir s\u00fcr\u00fcm\u00fcd\u00fcr ve sadece conda<\/code> paket y\u00f6neticisini ve Python’\u0131 i\u00e7erir.<\/p>\n
Miniconda Kurulumu<\/h4>\n
1.  Miniconda \u0130ndirme:<\/strong> Terminalinizde a\u015fa\u011f\u0131daki komutla en g\u00fcncel Miniconda y\u00fckleyicisini indirin:<\/p>\n
wget https:\/\/repo.anaconda.com\/miniconda\/Miniconda3-latest-Linux-x86_64.sh<\/code><\/pre>\n
2.  Miniconda Kurulumu:<\/strong> \u0130ndirilen beti\u011fi \u00e7al\u0131\u015ft\u0131r\u0131n:<\/p>\n
bash Miniconda3-latest-Linux-x86_64.sh<\/code><\/pre>\n
    Kurulum s\u0131ras\u0131nda lisans s\u00f6zle\u015fmesini onaylaman\u0131z (Enter tu\u015funa basarak okuyun, ard\u0131ndan yes<\/code> yaz\u0131n), kurulum dizinini onaylaman\u0131z (varsay\u0131lan\u0131 kabul etmek i\u00e7in Enter) ve conda init<\/code> komutunu \u00e7al\u0131\u015ft\u0131rmay\u0131 kabul etmeniz istenecektir (yes<\/code> yaz\u0131n).<\/p>\n
3.  Ortam De\u011fi\u015fikliklerini Etkinle\u015ftirme:<\/strong> Kurulum tamamland\u0131ktan sonra, terminal oturumunuzu yeniden ba\u015flatman\u0131z veya .bashrc<\/code> dosyas\u0131n\u0131 tekrar kaynaklaman\u0131z gerekir:<\/p>\n
source ~\/.bashrc<\/code><\/pre>\n
    Art\u0131k terminalinizde (base)<\/code> \u00f6n eki g\u00f6rmelisiniz, bu da base<\/code> conda ortam\u0131n\u0131n etkin oldu\u011funu g\u00f6sterir.<\/p>\n
Yeni Bir Conda Ortam\u0131 Olu\u015fturma<\/h4>\n
Projeniz i\u00e7in izole bir ortam olu\u015ftural\u0131m:<\/p>\n
conda create -n ai_env python=3.9 -y<\/code><\/pre>\n
Bu komut, ai_env<\/code> ad\u0131nda, Python 3.9 s\u00fcr\u00fcm\u00fcn\u00fc i\u00e7eren yeni bir conda ortam\u0131 olu\u015fturacakt\u0131r. -y<\/code> bayra\u011f\u0131, onay sorular\u0131n\u0131 otomatik olarak yes<\/code> olarak yan\u0131tlar.<\/p>\n
Conda Ortam\u0131n\u0131 Etkinle\u015ftirme<\/h4>\n
Olu\u015fturdu\u011funuz ortam\u0131 etkinle\u015ftirin:<\/p>\n
conda activate ai_env<\/code><\/pre>\n
Art\u0131k terminalinizde (ai_env)<\/code> \u00f6n ekini g\u00f6rmelisiniz. Bu, t\u00fcm y\u00fckleyece\u011finiz k\u00fct\u00fcphanelerin bu ortama kurulaca\u011f\u0131 anlam\u0131na gelir.<\/p>\n
AI\/ML K\u00fct\u00fcphanelerinin Kurulumu<\/h3>\n
\u015eimdi, etkinle\u015ftirdi\u011fimiz ai_env<\/code> ortam\u0131na pop\u00fcler AI\/ML k\u00fct\u00fcphanelerini kural\u0131m. GPU’yu kullanabilen s\u00fcr\u00fcmleri y\u00fcklemeye dikkat edin.<\/p>\n
PyTorch Kurulumu<\/h4>\n
PyTorch, esnekli\u011fi ve kolay kullan\u0131m\u0131 nedeniyle AI\/ML geli\u015ftiricileri aras\u0131nda olduk\u00e7a pop\u00fclerdir.<\/p>\n
1.  PyTorch Resmi Sitesine Gitme:<\/strong> pytorch.org<\/code> adresine gidin ve “Install PyTorch” b\u00f6l\u00fcm\u00fcne t\u0131klay\u0131n.
\n2.  S\u00fcr\u00fcm Se\u00e7imi:<\/strong> \u0130\u015fletim sistemi (Linux), paket y\u00f6neticisi (pip), Python s\u00fcr\u00fcm\u00fc (3.x) ve en \u00f6nemlisi CUDA s\u00fcr\u00fcm\u00fcn\u00fcz\u00fc se\u00e7in. \u00d6rne\u011fin, CUDA 12.1 veya 12.2 i\u00e7in cu121<\/code> veya cu12x<\/code> se\u00e7ene\u011fini tercih edin. Se\u00e7imlerinize g\u00f6re size \u00f6zel bir pip install<\/code> komutu olu\u015fturulacakt\u0131r.
\n3.  Kurulum Komutunu \u00c7al\u0131\u015ft\u0131rma (\u00f6rnek):<\/strong><\/p>\n
pip install torch torchvision torchaudio --index-url https:\/\/download.pytorch.org\/whl\/cu121<\/code><\/pre>\n
    Not:<\/em> Yukar\u0131daki komut \u00f6rnek bir CUDA s\u00fcr\u00fcm\u00fc i\u00e7ermektedir. Kendi CUDA Toolkit s\u00fcr\u00fcm\u00fcn\u00fcze uygun olan\u0131 PyTorch web sitesinden kopyalad\u0131\u011f\u0131n\u0131zdan emin olun.<\/p>\n
TensorFlow Kurulumu<\/h4>\n
TensorFlow, Google taraf\u0131ndan geli\u015ftirilen bir ba\u015fka pop\u00fcler AI\/ML k\u00fct\u00fcphanesidir.<\/p>\n
1.  TensorFlow Kurulumu:<\/strong> TensorFlow’un GPU destekli s\u00fcr\u00fcm\u00fcn\u00fc kurmak i\u00e7in (TensorFlow 2.x ve sonras\u0131 i\u00e7in genellikle CUDA ve cuDNN ba\u011f\u0131ml\u0131l\u0131klar\u0131n\u0131 otomatik olarak \u00e7\u00f6zer):<\/p>\n
pip install tensorflow[and-cuda] # TensorFlow 2.10 ve sonras\u0131 i\u00e7in\n    # Veya belirli bir s\u00fcr\u00fcm i\u00e7in:\n    # pip install tensorflow==2.15.0<\/code><\/pre>\n
    Not:<\/em> TensorFlow’un GPU deste\u011fi i\u00e7in CUDA Toolkit ve cuDNN’in do\u011fru bir \u015fekilde kurulmu\u015f olmas\u0131 gerekir. tensorflow[and-cuda]<\/code> paketi \u00e7o\u011fu ba\u011f\u0131ml\u0131l\u0131\u011f\u0131 otomatik olarak y\u00fckler, ancak bazen manuel m\u00fcdahale gerekebilir.<\/p>\n
Di\u011fer Temel K\u00fct\u00fcphaneler<\/h4>\n
AI\/ML projeleri i\u00e7in vazge\u00e7ilmez olan di\u011fer baz\u0131 k\u00fct\u00fcphaneleri de kural\u0131m:<\/p>\n
pip install numpy pandas scikit-learn matplotlib seaborn tqdm<\/code><\/pre>\n
GPU Alg\u0131lamas\u0131n\u0131 Do\u011frulama<\/h4>\n
K\u00fct\u00fcphanelerin GPU’nuzu ba\u015far\u0131yla kullan\u0131p kullanamad\u0131\u011f\u0131n\u0131 kontrol edelim:<\/p>\n
1.  Python Kabu\u011funa Girme:<\/strong><\/p>\n
python<\/code><\/pre>\n
2.  PyTorch i\u00e7in Do\u011frulama:<\/strong><\/p>\n
import torch\n    print(torch.cuda.is_available())\n    print(torch.cuda.device_count())\n    print(torch.cuda.get_device_name(0))<\/code><\/pre>\n
    True<\/code> \u00e7\u0131kt\u0131s\u0131 ve GPU ad\u0131n\u0131z\u0131 g\u00f6rmelisiniz.<\/p>\n
3.  TensorFlow i\u00e7in Do\u011frulama:<\/strong><\/p>\n
import tensorflow as tf\n    print(tf.config.list_physical_devices('GPU'))<\/code><\/pre>\n
    \u00c7\u0131kt\u0131da fiziksel GPU cihaz\u0131n\u0131z\u0131n listelendi\u011fini g\u00f6rmelisiniz.<\/p>\n
E\u011fer bu kontroller ba\u015far\u0131l\u0131 olursa, AI\/ML k\u00fct\u00fcphaneleriniz GPU’nuzu kullanmaya haz\u0131rd\u0131r.<\/p>\n
Jupyter Labs Kurulumu ve Yap\u0131land\u0131rmas\u0131<\/h3>\n
Jupyter Labs, AI\/ML geli\u015ftiricileri i\u00e7in etkile\u015fimli kodlama, veri analizi ve g\u00f6rselle\u015ftirme i\u00e7in vazge\u00e7ilmez bir web tabanl\u0131 geli\u015ftirme ortam\u0131d\u0131r. Droplet’imiz \u00fczerinde Jupyter Labs’\u0131 kurup g\u00fcvenli bir \u015fekilde eri\u015filebilir hale getirelim.<\/p>\n
Jupyter Labs Kurulumu<\/h4>\n
ai_env<\/code> conda ortam\u0131n\u0131z etkin durumdayken Jupyter Labs’\u0131 kurun:<\/p>\n
pip install jupyterlab<\/code><\/pre>\n
Jupyter Labs Yap\u0131land\u0131rmas\u0131<\/h4>\n
Jupyter Labs’\u0131 bir sunucu \u00fczerinde \u00e7al\u0131\u015ft\u0131rmak i\u00e7in baz\u0131 yap\u0131land\u0131rmalar yapmam\u0131z gerekiyor:<\/p>\n
1.  Yap\u0131land\u0131rma Dosyas\u0131 Olu\u015fturma:<\/strong><\/p>\n
jupyter lab --generate-config<\/code><\/pre>\n
    Bu komut, ~\/.jupyter\/jupyter_lab_config.py<\/code> yolunda bir yap\u0131land\u0131rma dosyas\u0131 olu\u015fturacakt\u0131r.<\/p>\n
2.  Jupyter Labs \u0130\u00e7in Parola Belirleme:<\/strong> Web aray\u00fcz\u00fcne eri\u015fim i\u00e7in bir parola belirlemek g\u00fcvenlik a\u00e7\u0131s\u0131ndan kritik \u00f6neme sahiptir.<\/p>\n
jupyter lab password<\/code><\/pre>\n
    Bu komut, parolay\u0131 \u015fifreleyip yap\u0131land\u0131rma dosyas\u0131na kaydedecektir.<\/p>\n
3.  Yap\u0131land\u0131rma Dosyas\u0131n\u0131 D\u00fczenleme:<\/strong> ~\/.jupyter\/jupyter_lab_config.py<\/code> dosyas\u0131n\u0131 bir metin d\u00fczenleyici ile a\u00e7\u0131n:<\/p>\n
nano ~\/.jupyter\/jupyter_lab_config.py<\/code><\/pre>\n
    A\u015fa\u011f\u0131daki sat\u0131rlar\u0131 bulun (veya ekleyin) ve de\u011ferlerini ayarlay\u0131n. Sat\u0131rlar\u0131n ba\u015f\u0131ndaki #<\/code> i\u015faretini kald\u0131rarak yorum sat\u0131r\u0131 olmaktan \u00e7\u0131kar\u0131n:<\/p>\n
c.ServerApp.ip = '0.0.0.0' # T\u00fcm a\u011f aray\u00fczlerinden eri\u015fime izin verir\n    c.ServerApp.open_browser = False # Sunucu ba\u015flat\u0131ld\u0131\u011f\u0131nda otomatik taray\u0131c\u0131 a\u00e7may\u0131 engeller\n    c.ServerApp.port = 8888 # Jupyter Labs'\u0131n dinleyece\u011fi port (varsay\u0131lan)<\/code><\/pre>\n
    Ctrl+X, Y, Enter tu\u015flar\u0131na basarak kaydedin ve \u00e7\u0131k\u0131n.<\/p>\n
Jupyter Labs’\u0131 Arka Planda \u00c7al\u0131\u015ft\u0131rma (Systemd Servisi ile)<\/h4>\n
Jupyter Labs’\u0131 bir systemd<\/code> servisi olarak \u00e7al\u0131\u015ft\u0131rmak, oturumunuzu kapatt\u0131\u011f\u0131n\u0131zda bile arka planda \u00e7al\u0131\u015fmaya devam etmesini sa\u011flar ve sunucunun yeniden ba\u015flat\u0131lmas\u0131 durumunda otomatik olarak ba\u015flamas\u0131na olanak tan\u0131r.<\/p>\n
1.  Systemd Servis Dosyas\u0131 Olu\u015fturma:<\/strong><\/p>\n
sudo nano \/etc\/systemd\/system\/jupyter.service<\/code><\/pre>\n
    Dosyan\u0131n i\u00e7ine a\u015fa\u011f\u0131daki i\u00e7eri\u011fi yap\u0131\u015ft\u0131r\u0131n. yourusername<\/code> ve ai_env<\/code> k\u0131s\u0131mlar\u0131n\u0131 kendi kullan\u0131c\u0131 ad\u0131n\u0131z ve conda ortam ad\u0131n\u0131zla de\u011fi\u015ftirin. ExecStart<\/code> sat\u0131r\u0131nda jupyter lab<\/code> komutunun tam yolunu belirtmek \u00f6nemlidir; bunu which jupyter-lab<\/code> komutuyla bulabilirsiniz (\u00f6rne\u011fin \/home\/yourusername\/miniconda3\/envs\/ai_env\/bin\/jupyter-lab<\/code>).<\/p>\n
[Unit]\n    Description=Jupyter Lab\n    After=network.target\n\n    [Service]\n    User=yourusername\n    Group=yourusername\n    WorkingDirectory=\/home\/yourusername\/\n    Environment=\"PATH=\/home\/yourusername\/miniconda3\/envs\/ai_env\/bin:$PATH\"\n    ExecStart=\/home\/yourusername\/miniconda3\/envs\/ai_env\/bin\/jupyter-lab --config=\/home\/yourusername\/.jupyter\/jupyter_lab_config.py\n    Restart=always\n\n    [Install]\n    WantedBy=multi-user.target<\/code><\/pre>\n
    Not:<\/em> WorkingDirectory<\/code> olarak \/home\/yourusername\/<\/code> belirledik. Bu, Jupyter Labs’\u0131n ana dizini olacakt\u0131r. Projelerinizi bu dizin alt\u0131nda tutabilirsiniz.<\/p>\n
    Kaydedin ve \u00e7\u0131k\u0131n.<\/p>\n
2.  Systemd Servisini Etkinle\u015ftirme ve Ba\u015flatma:<\/strong><\/p>\n
sudo systemctl daemon-reload # Yeni servis dosyas\u0131n\u0131 y\u00fckle\n    sudo systemctl enable jupyter # Servisi ba\u015flang\u0131\u00e7ta otomatik \u00e7al\u0131\u015facak \u015fekilde ayarla\n    sudo systemctl start jupyter # Servisi ba\u015flat<\/code><\/pre>\n
3.  Servis Durumunu Kontrol Etme:<\/strong><\/p>\n
sudo systemctl status jupyter<\/code><\/pre>\n
    active (running)<\/code> ibaresini g\u00f6rmelisiniz. Herhangi bir hata varsa, journalctl -u jupyter<\/code> komutuyla g\u00fcnl\u00fckleri kontrol edebilirsiniz.<\/p>\n
Jupyter Labs’a Eri\u015fim<\/h3>\n
Jupyter Labs servisi arka planda \u00e7al\u0131\u015ft\u0131\u011f\u0131na g\u00f6re, \u015fimdi web taray\u0131c\u0131n\u0131z \u00fczerinden eri\u015febiliriz.<\/p>\n
1.  G\u00fcvenlik Duvar\u0131 Kural\u0131 Ekleme:<\/strong> Jupyter Labs’\u0131n dinledi\u011fi porta (varsay\u0131lan 8888) g\u00fcvenlik duvar\u0131ndan izin vermemiz gerekiyor:<\/p>\n
sudo ufw allow 8888\/tcp<\/code><\/pre>\n
2.  Web Taray\u0131c\u0131s\u0131ndan Eri\u015fim:<\/strong> Web taray\u0131c\u0131n\u0131z\u0131 a\u00e7\u0131n ve a\u015fa\u011f\u0131daki adrese gidin:<\/p>\n
http:\/\/your_droplet_ip_address:8888<\/code><\/pre>\n
    your_droplet_ip_address<\/code> yerine droplet’inizin genel IP adresini yaz\u0131n.<\/p>\n
    Jupyter Labs’\u0131n parola ekran\u0131 ile kar\u015f\u0131la\u015facaks\u0131n\u0131z. Daha \u00f6nce belirledi\u011finiz parolay\u0131 girerek giri\u015f yap\u0131n. Art\u0131k AI\/ML kodlar\u0131n\u0131z\u0131 \u00e7al\u0131\u015ft\u0131rabilece\u011finiz tam i\u015flevsel bir Jupyter Labs ortam\u0131na sahipsiniz!<\/p>\n
En \u0130yi Uygulamalar ve Ek Hususlar<\/h3>\n
GPU Droplet’inizde AI\/ML ortam\u0131n\u0131z\u0131 kurdunuz. \u015eimdi bu ortam\u0131 daha verimli, g\u00fcvenli ve maliyet etkin kullanmak i\u00e7in baz\u0131 en iyi uygulamalara ve ek hususlara de\u011finelim.<\/p>\n
SSH T\u00fcnelleme (Daha G\u00fcvenli Eri\u015fim)<\/h4>\n
Jupyter Labs’a do\u011frudan IP adresi ve port \u00fczerinden eri\u015fmek yerine, SSH t\u00fcnelleme kullanarak daha g\u00fcvenli bir ba\u011flant\u0131 kurabilirsiniz. Bu y\u00f6ntem, Jupyter Labs portunu d\u0131\u015far\u0131ya a\u00e7madan, yerel makinenizdeki bir portu droplet’inizdeki Jupyter Labs portuna y\u00f6nlendirir.<\/p>\n
Terminalinizde (yerel makinenizde) a\u015fa\u011f\u0131daki komutu \u00e7al\u0131\u015ft\u0131r\u0131n:<\/p>\n
ssh -L 8888:localhost:8888 yourusername@your_droplet_ip_address<\/code><\/pre>\n
Bu komut, yerel makinenizin 8888 portunu droplet’inizin 8888 portuna t\u00fcneller. Bu durumda, droplet’inizde sudo ufw allow 8888\/tcp<\/code> komutunu \u00e7al\u0131\u015ft\u0131rman\u0131za gerek kalmaz. Taray\u0131c\u0131n\u0131zda http:\/\/localhost:8888<\/code> adresine giderek Jupyter Labs’a eri\u015febilirsiniz. SSH oturumu a\u00e7\u0131k kald\u0131\u011f\u0131 s\u00fcrece t\u00fcnel de a\u00e7\u0131k kal\u0131r.<\/p>\n
Snapshot’lar ve Yedeklemeler<\/h4>\n
\u00c7al\u0131\u015fma ortam\u0131n\u0131z\u0131 ve verilerinizi yedeklemek kritik \u00f6neme sahiptir. DigitalOcean, droplet’lerinizin anl\u0131k g\u00f6r\u00fcnt\u00fclerini (snapshots) alman\u0131z\u0131 sa\u011flar. Bu snapshot’lar, droplet’inizin o anki durumunun bir kopyas\u0131d\u0131r ve gelecekte yeni bir droplet olu\u015fturmak veya mevcut droplet’inizi \u00f6nceki bir duruma geri d\u00f6nd\u00fcrmek i\u00e7in kullan\u0131labilir.<\/p>\n
*   Ne Zaman Kullan\u0131l\u0131r:<\/strong> \u00d6nemli bir yap\u0131land\u0131rma de\u011fi\u015fikli\u011fi yapmadan \u00f6nce, b\u00fcy\u00fck bir proje \u00fczerinde \u00e7al\u0131\u015fmaya ba\u015flamadan \u00f6nce veya ortam\u0131n\u0131z\u0131 ba\u015fka bir droplet’e ta\u015f\u0131mak istedi\u011finizde.
\n*   Nas\u0131l Yap\u0131l\u0131r:<\/strong> DigitalOcean kontrol panelinde droplet’inizin sayfas\u0131na gidin, “Snapshots” sekmesine t\u0131klay\u0131n ve “Take Snapshot” butonuna bas\u0131n.<\/p>\n
Maliyet Y\u00f6netimi<\/h4>\n
GPU Droplet’leri, CPU tabanl\u0131 droplet’lerden daha pahal\u0131d\u0131r. Maliyetleri kontrol alt\u0131nda tutmak i\u00e7in:<\/p>\n
*   Kullan\u0131lmad\u0131\u011f\u0131nda Kapatma:<\/strong> E\u011fer uzun s\u00fcre kullanmayacaksan\u0131z, droplet’inizi kapatabilirsiniz. Ancak, DigitalOcean’\u0131n GPU Droplet’leri genellikle kapal\u0131yken bile GPU kaynaklar\u0131 i\u00e7in \u00fccretlendirilmeye devam eder. Bu nedenle, uzun s\u00fcreli aralar i\u00e7in droplet’i tamamen silip snapshot’tan geri y\u00fcklemeyi d\u00fc\u015f\u00fcnebilirsiniz. Fiyatland\u0131rma detaylar\u0131n\u0131 DigitalOcean’\u0131n resmi sayfas\u0131ndan kontrol edin.
\n*   \u0130htiyaca G\u00f6re \u00d6l\u00e7eklendirme:<\/strong> Projenizin gereksinimleri de\u011fi\u015ftik\u00e7e droplet plan\u0131n\u0131z\u0131 y\u00fckseltip al\u00e7altmay\u0131 d\u00fc\u015f\u00fcn\u00fcn.
\n*   Kullan\u0131m Takibi:<\/strong> DigitalOcean’\u0131n izleme (monitoring) ara\u00e7lar\u0131n\u0131 kullanarak GPU ve CPU kullan\u0131m\u0131n\u0131z\u0131 takip edin.<\/p>\n
Veri Aktar\u0131m\u0131<\/h4>\n
Yerel makineniz ile droplet’iniz aras\u0131nda veri aktarmak i\u00e7in scp<\/code> veya rsync<\/code> gibi ara\u00e7lar\u0131 kullanabilirsiniz:<\/p>\n
*   SCP (Secure Copy Protocol):<\/strong> Dosyalar\u0131 kopyalamak i\u00e7in basittir.
\n    *   Yerelden Droplet’e: scp \/path\/to\/local\/file.txt yourusername@your_droplet_ip_address:\/path\/to\/remote\/directory\/<\/code>
\n    *   Droplet’ten Yerele: scp yourusername@your_droplet_ip_address:\/path\/to\/remote\/file.txt \/path\/to\/local\/directory\/<\/code>
\n*   Rsync:<\/strong> B\u00fcy\u00fck dosyalar veya dizinler i\u00e7in daha verimlidir, sadece de\u011fi\u015fen k\u0131s\u0131mlar\u0131 kopyalar.
\n    *   Yerelden Droplet’e: rsync -avz \/path\/to\/local\/folder\/ yourusername@your_droplet_ip_address:\/path\/to\/remote\/folder\/<\/code><\/p>\n
Versiyon Kontrol\u00fc (Git)<\/h4>\n
Kodlar\u0131n\u0131z\u0131 ve projelerinizi y\u00f6netmek i\u00e7in Git kullanmak vazge\u00e7ilmezdir. Droplet’inizde Git’i kurarak GitHub, GitLab veya Bitbucket gibi platformlarla entegre olabilirsiniz:<\/p>\n
sudo apt install git -y<\/code><\/pre>\n
Ard\u0131ndan, projenizi klonlayabilir veya yeni bir Git deposu ba\u015flatabilirsiniz.<\/p>\n
Ortam \u0130zleme<\/h4>\n
DigitalOcean, droplet’lerinizin CPU, bellek, disk ve a\u011f kullan\u0131m\u0131n\u0131 izlemek i\u00e7in yerle\u015fik ara\u00e7lar sunar. Bu metrikleri kontrol panelinizden takip ederek performans darbo\u011fazlar\u0131n\u0131 veya a\u015f\u0131r\u0131 kaynak kullan\u0131m\u0131n\u0131 tespit edebilirsiniz. Ayr\u0131ca htop<\/code> (CPU ve bellek), iotop<\/code> (disk I\/O) ve nmon<\/code> (genel sistem izleme) gibi terminal tabanl\u0131 ara\u00e7lar\u0131 da y\u00fckleyerek sistem performans\u0131n\u0131z\u0131 daha detayl\u0131 izleyebilirsiniz.<\/p>\n
sudo apt install htop iotop nmon -y<\/code><\/pre>\n
Bu en iyi uygulamalar ve ek hususlar, AI\/ML geli\u015ftirme ortam\u0131n\u0131z\u0131 daha g\u00fcvenli, verimli ve s\u00fcrd\u00fcr\u00fclebilir hale getirmenize yard\u0131mc\u0131 olacakt\u0131r.<\/p>\n
Sonu\u00e7<\/h3>\n
Bu kapsaml\u0131 rehberde, yapay zeka ve makine \u00f6\u011frenimi projeleriniz i\u00e7in DigitalOcean GPU Droplet \u00fczerinde g\u00fc\u00e7l\u00fc ve esnek bir geli\u015ftirme ortam\u0131n\u0131 ad\u0131m ad\u0131m kurmay\u0131 \u00f6\u011frendiniz. Ba\u015flang\u0131\u00e7ta bir GPU Droplet’in nas\u0131l olu\u015fturulaca\u011f\u0131ndan, temel sunucu g\u00fcvenli\u011fi ve yap\u0131land\u0131rmas\u0131na; NVIDIA s\u00fcr\u00fcc\u00fcleri ve CUDA Toolkit’in kritik kurulumundan, Python ortam\u0131n\u0131n Miniconda ile izole edilmesine; pop\u00fcler AI\/ML k\u00fct\u00fcphanelerinin (PyTorch, TensorFlow) y\u00fcklenmesinden, Jupyter Labs’\u0131n kurulup g\u00fcvenli bir \u015fekilde eri\u015filebilir hale getirilmesine kadar t\u00fcm s\u00fcreci detayl\u0131 bir \u015fekilde ele ald\u0131k.<\/p>\n
Art\u0131k, yerel donan\u0131m s\u0131n\u0131rlamalar\u0131n\u0131n \u00f6tesine ge\u00e7erek, bulutun g\u00fcc\u00fcn\u00fc kullanarak AI\/ML modellerinizi e\u011fitebilir, b\u00fcy\u00fck veri setlerini analiz edebilir ve interaktif bir ortamda kod geli\u015ftirebilirsiniz. Bu kurulum, hem yeni ba\u015flayanlar hem de deneyimli geli\u015ftiriciler i\u00e7in h\u0131zl\u0131 prototipleme ve proje geli\u015ftirme imkanlar\u0131 sunar. Ayr\u0131ca, SSH t\u00fcnelleme, snapshot’lar, maliyet y\u00f6netimi ve veri aktar\u0131m\u0131 gibi en iyi uygulamalarla ortam\u0131n\u0131z\u0131 daha g\u00fcvenli, verimli ve s\u00fcrd\u00fcr\u00fclebilir hale getirme konusunda da bilgi edindiniz.<\/p>\n
Bu ortam, AI\/ML yolculu\u011funuzda size sa\u011flam bir temel sa\u011flayacakt\u0131r. Art\u0131k GPU h\u0131zland\u0131rmal\u0131 hesaplama yetenekleriyle donat\u0131lm\u0131\u015f bir geli\u015ftirme ortam\u0131na sahipsiniz. Jupyter Labs’ta not defterleri olu\u015fturarak, veri bilimi ve makine \u00f6\u011frenimi projelerinizi hayata ge\u00e7irmeye ba\u015flayabilirsiniz. Unutmay\u0131n, \u00f6\u011frenme ve ke\u015ffetme s\u00fcreci devaml\u0131d\u0131r. Bu ortam\u0131 kendi \u00f6zel ihtiya\u00e7lar\u0131n\u0131za g\u00f6re daha da optimize etmek ve farkl\u0131 ara\u00e7lar\u0131 entegre etmek i\u00e7in deneyler yapmaktan \u00e7ekinmeyin. Ba\u015far\u0131lar dileriz!<\/body><\/p>\n","protected":false},"excerpt":{"rendered":"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7\nYapay zeka (AI) ve makine \u00f6\u011frenimi (ML)","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"csco_page_header_type":"","csco_page_load_nextpost":"","csco_page_subscribe_form":"","csco_page_contact_form":"","footnotes":""},"categories":[1],"tags":[],"class_list":{"0":"post-31607","1":"post","2":"type-post","3":"status-publish","4":"format-standard","6":"category-genel","7":"cs-entry","8":"cs-video-wrap"},"yoast_head":"\nYapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7 - Kodlar\u0131n Gizemli D\u00fcnyas\u0131<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/\" \/>\n<meta property=\"og:locale\" content=\"tr_TR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7\" \/>\n<meta property=\"og:description\" content=\"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7 Yapay zeka (AI) ve makine \u00f6\u011frenimi (ML)\" \/>\n<meta property=\"og:url\" content=\"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/\" \/>\n<meta property=\"og:site_name\" content=\"Kodlar\u0131n Gizemli D\u00fcnyas\u0131\" \/>\n<meta property=\"article:published_time\" content=\"2025-10-11T18:41:17+00:00\" \/>\n<meta name=\"author\" content=\"Fatih Soysal\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Yazan:\" \/>\n\t<meta name=\"twitter:data1\" content=\"Fatih Soysal\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tahmini okuma s\u00fcresi\" \/>\n\t<meta name=\"twitter:data2\" content=\"24 dakika\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/\"},\"author\":{\"name\":\"Fatih Soysal\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"headline\":\"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7\",\"datePublished\":\"2025-10-11T18:41:17+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/\"},\"wordCount\":4170,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"inLanguage\":\"tr\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/#respond\"]}],\"copyrightYear\":\"2025\",\"copyrightHolder\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/\",\"name\":\"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7 - Kodlar\u0131n Gizemli D\u00fcnyas\u0131\",\"isPartOf\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#website\"},\"datePublished\":\"2025-10-11T18:41:17+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/#breadcrumb\"},\"inLanguage\":\"tr\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Anasayfa\",\"item\":\"https:\/\/fatihsoysal.com\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#website\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/\",\"name\":\"Fatihsoysal.com\",\"description\":\"Blog - Yaz\u0131l\u0131m D\u00fcnyas\u0131 Tecr\u00fcbelerim\",\"publisher\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/fatihsoysal.com\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"tr\"},{\"@type\":[\"Person\",\"Organization\"],\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\",\"name\":\"Fatih Soysal\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"tr\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png\",\"contentUrl\":\"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png\",\"width\":512,\"height\":512,\"caption\":\"Fatih Soysal\"},\"logo\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/\"},\"description\":\"Kullan\u0131m ve kodlama m\u00fckemmeliyetini odak alan uygulamalar olu\u015fturma deneyimine sahip, profesyonel olarak 15+ y\u0131l \u00fczeri deneyime sahip bir yaz\u0131l\u0131m m\u00fchendisi.\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/author\/fatihsoysal\/\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7 - Kodlar\u0131n Gizemli D\u00fcnyas\u0131","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/","og_locale":"tr_TR","og_type":"article","og_title":"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7","og_description":"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7 Yapay zeka (AI) ve makine \u00f6\u011frenimi (ML)","og_url":"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/","og_site_name":"Kodlar\u0131n Gizemli D\u00fcnyas\u0131","article_published_time":"2025-10-11T18:41:17+00:00","author":"Fatih Soysal","twitter_card":"summary_large_image","twitter_misc":{"Yazan:":"Fatih Soysal","Tahmini okuma s\u00fcresi":"24 dakika"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/#article","isPartOf":{"@id":"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/"},"author":{"name":"Fatih Soysal","@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"headline":"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7","datePublished":"2025-10-11T18:41:17+00:00","mainEntityOfPage":{"@id":"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/"},"wordCount":4170,"commentCount":0,"publisher":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"inLanguage":"tr","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/#respond"]}],"copyrightYear":"2025","copyrightHolder":{"@id":"https:\/\/fatihsoysal.com\/blog\/#organization"}},{"@type":"WebPage","@id":"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/","url":"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/","name":"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7 - Kodlar\u0131n Gizemli D\u00fcnyas\u0131","isPartOf":{"@id":"https:\/\/fatihsoysal.com\/blog\/#website"},"datePublished":"2025-10-11T18:41:17+00:00","breadcrumb":{"@id":"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/#breadcrumb"},"inLanguage":"tr","potentialAction":[{"@type":"ReadAction","target":["https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/fatihsoysal.com\/blog\/yapay-zeka-ve-makine-ogrenimi-icin-gpu-droplet-ortami-kurulumu-jupyter-labs-ile-kodlamaya-baslangic\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Anasayfa","item":"https:\/\/fatihsoysal.com\/blog\/"},{"@type":"ListItem","position":2,"name":"Yapay Zeka ve Makine \u00d6\u011frenimi i\u00e7in GPU Droplet Ortam\u0131 Kurulumu: Jupyter Labs ile Kodlamaya Ba\u015flang\u0131\u00e7"}]},{"@type":"WebSite","@id":"https:\/\/fatihsoysal.com\/blog\/#website","url":"https:\/\/fatihsoysal.com\/blog\/","name":"Fatihsoysal.com","description":"Blog - Yaz\u0131l\u0131m D\u00fcnyas\u0131 Tecr\u00fcbelerim","publisher":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/fatihsoysal.com\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"tr"},{"@type":["Person","Organization"],"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1","name":"Fatih Soysal","image":{"@type":"ImageObject","inLanguage":"tr","@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/","url":"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png","contentUrl":"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png","width":512,"height":512,"caption":"Fatih Soysal"},"logo":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/"},"description":"Kullan\u0131m ve kodlama m\u00fckemmeliyetini odak alan uygulamalar olu\u015fturma deneyimine sahip, profesyonel olarak 15+ y\u0131l \u00fczeri deneyime sahip bir yaz\u0131l\u0131m m\u00fchendisi.","url":"https:\/\/fatihsoysal.com\/blog\/author\/fatihsoysal\/"}]}},"yoast_meta":{"yoast_wpseo_title":"","yoast_wpseo_metadesc":"","yoast_wpseo_canonical":""},"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/31607","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/comments?post=31607"}],"version-history":[{"count":1,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/31607\/revisions"}],"predecessor-version":[{"id":31608,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/31607\/revisions\/31608"}],"wp:attachment":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/media?parent=31607"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/categories?post=31607"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/tags?post=31607"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}