Foundations — Transpose — definition, properties

2.6.6 · D1 · Maths › Matrices & Determinants — Introduction › Transpose — definition, properties

Parent note ki ek bhi line par trust karne se pehle, aapko usme aane wale har symbol ko khud se samajhna hoga. Yeh page unhe ek-ek karke build karta hai, numbers ki grid ke basic idea se lekar multiplication aur inverses ki machinery tak. Yahan kuch bhi assume nahi kiya gaya ki aapne pehle matrices dekhi hain.

1. Matrix kya hoti hai? (the grid)

Neeche diye figure ko dekhein. Baayein taraf ki grid ek matrix hai. Across padhna ek row hai; neeche padhna ek column hai.

Topic ko yeh kyun chahiye: transpose ek operation hai jo poori grid par ek saath hota hai. Agar aap matrix ko physical grid of pigeonholes ki tarah picture nahi karte, toh flip karne ke liye kuch hoga hi nahi. Baad mein aap Matrix Operations se milenge jo inhi grids par kaam karte hain.

2. Size: $m\times n$ (grid kitni badi hai?)

Topic ko yeh kyun chahiye: transpose size ko swap karta hai $m\times n$ se $n\times m$ mein. Parent note kehta hai "$A^T$ ek $n\times m$ matrix hai" — do numbers ka woh swap poori geometric story hai, aur aap ise tab hi dekh sakte hain jab aapko pata ho ki do numbers ka matlab kya hai.

3. Entries aur index $a_{ij}$ (ek single box ka naam)

Topic ko yeh kyun chahiye: transpose ki poori definition index language mein likhi jaati hai: $[A^T{]}_{ij}=[A{]}_{ji}$ Yeh formula itna hi kehta hai ki "flipped grid ke address $(i,j)$ par entry dhundne ke liye, original grid ke swapped address $(j,i)$ par jaao." Agar $a_{ij}$ aapke liye mystery hai, toh yeh formula padha hi nahi jaayega. Address master karo aur formula obvious ho jaayega.

4. Main diagonal (the mirror line)

Topic ko yeh kyun chahiye: isi se symmetric aur skew-symmetric matrices possible hoti hain.

• Ek symmetric matrix ($A^T=A$) woh hai jahan mirror image original ke identical dikhti hai — kuch nahi hila dikhta.
• Ek skew-symmetric matrix ($A^T=-A$) woh hai jahan mirror sign flip karta hai. Kyunki ek diagonal entry apne aap mein mirror hoti hai, hameein chahiye $a_{ii}=-a_{ii}$, jo $a_{ii}=0$ force karta hai — isi liye parent note insist karta hai ki ek skew-symmetric matrix ki diagonal sab zeros hoti hai. Symmetric Matrices dekhein.

5. Square vs rectangular

Sirf square matrices ka hi ek main diagonal hota hai jo corner se corner tak chalta hai, isliye symmetry, Determinants, aur Matrix Inverse sab square-matrix land mein rehte hain. Ek rectangular matrix jaise $2\times 3$ ko transpose kiya ja sakta hai (woh $3\times 2$ ban jaati hai), lekin woh kabhi apne transpose ke barabar nahi ho sakti — sizes match nahi karti, entries ki toh baat hi chhodo.

6. Row aur column vectors (woh strips jo flip hoti hain)

Topic ko yeh kyun chahiye: poora "flip rows into columns" description inhi single-strip moves se bana hai. Yeh Inner Product Spaces se bhi connect hota hai, jahan ek row-vector-times-column-vector se dot product likha jaata hai.

7. Matrix addition aur scalar multiples (asaan operations)

Topic ko yeh kyun chahiye: chaar "SIRP" properties mein se do inhi ke baare mein hain.

• $(A+B{)}^T=A^T+B^T$: addition entry-by-entry hoti hai, aur flipping address-by-address hota hai, isliye dono operations ek doosre mein interfere nahi karte — order matter nahi karta.
• $(kA{)}^T=k{A}^T$: ek scalar har box ko equally touch karta hai, isliye mirror nahi bata sakta ki scaling pehle ki gayi thi ya baad mein. Yeh asaan isliye hain (koi reversal nahi!) kyunki addition aur scaling har box ko independently treat karte hain.

8. Matrix multiplication aur sum $\sum$ (mushkil operation)

Yeh genuinely nayi machine hai, aur reversal rule poori tarah isi par depend karta hai.

Topic ko yeh kyun chahiye — reversal zero se explain kiya gaya: $(AB{)}^T$ mein hum address $(i,j)$ maangte hain, jo $(AB)$ ka address $(j,i)$ hai, yaani ${\sum }_k{a}_{jk}{b}_{ki}$. Jab hum har factor ko alag se transpose karte hain, toh har $a_{jk}$ swapped address par $A^T$ ki entry ban jaata hai, aur $B$ ke saath bhi yahi hota hai. Shared index $k$ ko do matrices ke beech mein wapas line up karne ka ek hi tarika hai — $B^T$ ko pehle aur $A^T$ ko doosre place par rakhna. Yeh koi trick nahi hai — yeh forced hai is baat se ki kaunsi finger kisse milni hai. Isi liye $(AB{)}^T=B^T{A}^T$ hota hai na ki $A^T{B}^T$. Yeh machinery Eigenvalues and Eigenvectors aur Orthogonal Matrices mein bhi kaam aati hai.

9. Identity $I$ aur inverse $A^{-1}$

Topic ko yeh kyun chahiye: property $(A^{-1}{)}^T=(A^T{)}^{-1}$ equation $A{A}^{-1}=I$ ko transpose karke, reversal rule use karke, aur iss fact se prove ki jaati hai ki $I^T=I$ (identity symmetric hai — uski mirror image wahi hai). Yeh Orthogonal Matrices ki bhi key hai, jahan $A^T=A^{-1}$. Poori details Matrix Inverse mein hain.

Foundations topic ko kaise feed karte hain

Ise top-down padho: plain grid aapko size aur addresses deti hai; addresses plus diagonal aapko transpose deta hai; transpose plus har operation aapko har property deta hai; aur sab kuch symmetric, inverse, aur orthogonal matrices par converge hota hai — parent topic Transpose — definition, properties ka payoff.

Equipment checklist

Khud test karo — reveal karne se pehle har jawab zor se bolo.