Agar aap ek hi code ko baar-baar copy-paste kar rahe hain, toh aap ek "Coder" nahi hain. Ek professional developer Functions banata hai. Function ka matlab hai — ek baar code likho, aur jab chahe use "Call" karo. AI projects mein functions aapke code ko chhota, saaf, aur "Reusable" banate hain. Ise hum DRY (Don't Repeat Yourself) principle kehte hain.
1. The Function Logic: Input-Process-Output
Function ko ek Juicer ki tarah samjhein:
- Input (Arguments): Jo aapne machine ko diya (Fruits).
- Process: Jo machine ke andar hua (Blending).
- Output (Return): Jo aapko wapas mila (Juice). AI mein hum functions use karte hain data ko transform karne ya model se prediction nikalne ke liye.
2. Advanced Parameters: *args aur **kwargs
Kayi baar humein nahi pata hota ki user kitne inputs dega (e.g. Model ki configuration).
*args(Variable Arguments): Ye unlimited inputs ko ek Tuple mein pack kar deta hai.**kwargs(Keyword Arguments): Ye unlimited inputs ko ek Dictionary mein pack kar deta hai.- Professional AI libraries jaise PyTorch inhein har jagah use karti hain taaki aap flexible tareeqe se parameters pass kar sakein.
3. Lambda Functions: One-line Magic
Chote kaamo ke liye poora function banana tedious hota hai.
- Lambda: Ye "Anonymous" (bin-naam wale) functions hote hain.
square = lambda x: x * x- AI mein jab hum data cleaning (
mapyaapply) karte hain, toh Lambda hamara ghanton ka kaam seconds mein kar deta hai.
4. Scopes: Local vs Global (Dimaag ka daira)
- Local Scope: Jo variable function ke "Andar" banta hai, wo bahar nahi dikhta. (e.g. Temporary calculations).
- Global Scope: Jo variable bahar banta hai, wo har jagah dikhta hai. (e.g. Model weights). AI models mein hum hamesha variables ke scope ka dhyan rakhte hain taaki memory barbad na ho.
5. Summary Table: Function Elements for AI
| Element | Purpose | Why in AI? |
|---|---|---|
| def | Defining the block | Reusable model layers |
| return | Sending result | Getting predictions |
| default args | Pre-set values | Default learning rates |
| docstrings | Documentation | Explaining model logic |
FAQs
1. Print aur Return mein kya fark hai?
print sirf screen par dikhata hai (for humans). return us result ko "Variable" mein store karne ke layak banata hai taaki use aage math mein use kiya ja sake. AI pipelines mein hum hamesha return use karte hain.
2. Python mein "Recursive Function" kya hota hai? Aisa function jo apne aap ko hi call kare. Ye Tree-based algorithms (jaise Decision Trees) build karte waqt bahut kaam aata hai.
3. Type Hinting kyon zaroori hai?
def add(x: int, y: int) -> int: likhne se doosre developers (aur AI tools) ko pata chalta hai ki ye function kya umeed kar raha hai. 2026 mein pro code hamesha type-hinted hota hai.
4. "Pass-by-Reference" ya "Pass-by-Value"? Python mein sab kuch "Objects" hai. Agar aap list pass karte hain, toh wo original list ko change kar sakta hai. Ise dhyan mein rakhna zaroori hai jab aap data cleaning kar rahe hon.
Functions code ko "Professional" banate hain. Inhein master karke aap ek senior developer ki tarah AI code likhna shuru kar denge! 🚀
Tarun ke baare mein: Tarun modular programming aur clean-code architecture ke specialist hain. AI-Gyani par har function reusable aur efficient hai.