Lesson · Hinglish · ⏱ 8 min padhne mein

🐼 Pandas: data ki tables ke saath kaam karna

Q: Aapko har city ki total sales chahiye. Kaunsa Pandas tool sahi hai?

groupby(). groupby rows ko city ke hisaab se baantta hai, phir sum() har group ki sales jodta hai.

🏠 Asli zindagi ki misaal

Agar aapne Excel use kiya hai, to idea aapko pehle se pata hai: rows, columns, filters aur totals. Pandas Python ke andar Excel hai. Fark yeh hai ki yeh lakhon rows handle kar leta hai, badi files pe crash nahi hota, aur har baar aapke steps bilkul same tarah repeat karta hai.

💡 Aasaan shabdon mein

Pandas tabular data ke liye standard Python library hai. DataFrame ek table hai, aur uska har column ek Series hai. Kuch hi lines mein aap CSV load kar sakte ho, pehli rows dekh sakte ho, filter aur sort kar sakte ho, missing values bhar sakte ho, aur data group karke totals ya averages nikaal sakte ho. Koi bhi ML model seekhne se pehle data lagbhag hamesha Pandas mein hi clean aur shape kiya jaata hai. NumPy, jis pe Pandas bana hai, numbers ke arrays pe fast maths karta hai.

🪜 Step by step

Data load karo: df = pd.read_csv("sales.csv"). DataFrame ek table hai: rows aur naam wale columns.
Analyse karne se pehle dekho: df.head(), df.shape, df.info() (types aur missing values), df.describe() (summary numbers).
Select aur filter: df["city"], df[df["amount"] > 1000], df[["city", "amount"]].
Saaf karo: df.dropna(), df.fillna(0), df.drop_duplicates(), pd.to_datetime(df["date"]).
Summarise: df.groupby("city")["amount"].sum().sort_values(ascending=False). Yeh ek line kai business sawaalon ka jawaab deti hai.

🔑 Zaroori shabd

DataFrame: Rows aur naam wale columns wali table.
Missing value (NaN): Khaali cell. Modelling se pehle isse handle karna zaroori hai.
groupby: Rows ko groups mein baanto aur har group ka summary nikaalo, jaise har city ki sales.

👀 Example dekho

python

import pandas as pd

df = pd.read_csv("orders.csv")     # city, item, amount
print(df.head())                     # first 5 rows
df = df.dropna(subset=["amount"])   # drop rows with no amount
by_city = df.groupby("city")["amount"].sum().sort_values(ascending=False)
print(by_city.head(3))               # top 3 cities by sales

Load → clean → summarise. Data kaam ka roz ka loop.

⚠️ Aam galtiyaan jinse bachna hai

Data types check na karna: text mein save numbers ("1,200") jud nahi paayenge. pd.to_numeric se saaf karke convert karo.
for se row by row loop chalana. Pandas mein poore column pe operations bahut tez hote hain.
Bina soche missing values drop karna. Aap zyaadatar data mita sakte ho ya bias daal sakte ho; pehle dekho kitne hain aur kyun.

✅ Jaldi se check karo

Aapko har city ki total sales chahiye. Kaunsa Pandas tool sahi hai?

groupby()
read_csv()
head()

Jawaab dekho

A. groupby(): groupby rows ko city ke hisaab se baantta hai, phir sum() har group ki sales jodta hai.

🙋 Seekhne wale yeh poochhte hain

Pandas ya SQL: kaunsa use karun?

Dono. Database ke andar data nikalne aur aggregate karne ke liye SQL, phir Python mein exploration, cleaning aur ML prep ke liye pandas.

Meri file pandas ke liye bahut badi hai. Ab kya?

Sirf zaroori columns padho (usecols), chunks mein padho (chunksize), ya bade data ke tools jaise Polars, DuckDB ya Spark use karo.

📌 Yaad rakho

DataFrame = table. Zyaadatar AI projects load → clean → groupby se shuru hote hain.

Poora course, aapke roadmap ke hisaab se

Progress tracking, quick checks aur step-by-step order ke saath. Yeh lesson in courses ka hissa hai:

AI Foundations Path · ML Engineer · Data Engineer (AI focus)

Course mein seekho