📚 Playlist Notes — Index & Roadmap¶

"Numpy + Pandas + Matplotlib + Seaborn — CampusX" (15 videos)¶

Ye notes kaise use karein¶

Ye notes CampusX ke DSMP (Data Science Mentorship Program) sessions ke hisaab se organize kiye gaye hain — wahi order jo playlist follow karti hai. Har file self-contained hai, code ke saath.

Revision strategy: 1. Pehle Quick Revision table dekho (har file ke end mein hai) 2. Jo topic bhool gaye ho, uska section khol ke code padho 3. Colab kholo aur code khud chala ke dekho — sirf padhne se yaad nahi rehta

📁 Files¶

File	Kya hai andar	Sessions
01_NumPy_Complete_Notes.md	Array creation, attributes, indexing/slicing, operations, functions, reshaping, stacking, splitting, fancy & boolean indexing, broadcasting, ML formulas, missing values, 30+ NumPy tricks	Session 13, 14, 15
02_Pandas_Series_DataFrame.md	Series banana/methods/math/boolean indexing, DataFrame creation, iloc/loc, filtering, naye columns, 19 important DataFrame methods	Session 16, 17, 18
03_Advanced_Pandas.md	GroupBy (split-apply-combine), merge/concat/join, MultiIndex, stack/unstack, melt, pivot_table, `.str` vectorized string ops, DateTime & `.dt`	Session 19, 20, 21, 22
04_Matplotlib_Complete.md	Line, scatter, bar, histogram, pie, subplots, 3D plots, contour, heatmap, annotations, styles, Pandas plotting	Matplotlib sessions
05_Seaborn_Complete.md	Figure vs axes level, relational/distribution/categorical/regression/matrix/multiplot, pairplot, jointplot, heatmap, themes, full EDA workflow	Seaborn sessions

🗺️ Playlist ka Structure (CampusX DSMP order)¶

Week 5 — NumPy¶

#	Session	File
1	Session 13 — Numpy Fundamentals	`01_NumPy` → Session 13
2	Session 14 — Advanced Numpy	`01_NumPy` → Session 14
3	Session 15 — Numpy Tricks	`01_NumPy` → Session 15

Week 6 — Pandas¶

#	Session	File
4	Session 16 — Pandas Series	`02_Pandas` → Session 16
5	Important Series Methods (Supplementary)	`02_Pandas` → Section 10
6	Session 17 — Pandas DataFrame	`02_Pandas` → Session 17
7	Session 18 — Important DataFrame Methods	`02_Pandas` → Session 18

Week 7 — Advanced Pandas¶

#	Session	File
8	Session 19 — GroupBy Object	`03_Advanced` → Session 19
9	Session 20 — Merging, Joining, Concatenating	`03_Advanced` → Session 20
10	Session 21 — MultiIndex Series & DataFrames	`03_Advanced` → Session 21
11	Session 22 — Vectorized String Ops + DateTime	`03_Advanced` → Session 22

Visualization¶

#	Session	File
12	Plotting using Matplotlib (Part 1)	`04_Matplotlib` → Part 1-5
13	Advanced Matplotlib (Part 2)	`04_Matplotlib` → Part 6-11
14	Plotting using Seaborn (Part 1)	`05_Seaborn` → Part 1-3
15	Advanced Seaborn (Part 2)	`05_Seaborn` → Part 4-8

📦 Datasets jo CampusX ke Colab notebooks mein use hote hain¶

Agar tumhe khud practice karni hai toh ye datasets chahiye honge:

Dataset	Kahan use hota hai
`kohli_ipl.csv`	Series (Session 16)
`subs.csv`	Series — YouTube subscribers
`bollywood.csv`	Series — movie → actor
`movies.csv` / `imdb-top-1000.csv`	DataFrame, GroupBy
`ipl-matches.csv`	DataFrame filtering, value_counts
`titanic.csv`	String ops, EDA
`batsman_runs_ipl.csv`	rank, sort
`courses.csv`, `students.csv`, `nov.csv`, `dec.csv`, `regs.csv`	Merging (Session 20)
`matches.csv`, `deliveries.csv`	Merging practice questions
`time_series_covid19_*.csv`	melt (Session 21)
`expense_data.csv`	pivot_table
`sharma-kohli.csv`, `batter.csv`, `vk.csv`, `gayle-175.csv`	Matplotlib
Seaborn built-in: `tips`, `iris`, `titanic`, `flights`	Seaborn

Zyadatar CampusX datasets yahan milte hain: GitHub par campusx-official organization mein, ya har session ke YouTube description mein diye gaye Colab notebook mein.

⚡ Setup — har notebook ke shuru mein¶

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

pd.set_option('display.max_columns', None)     # saare columns dikhao
pd.set_option('display.max_rows', 100)
pd.set_option('display.float_format', '{:.2f}'.format)

%matplotlib inline
sns.set_theme(style='whitegrid')
plt.rcParams['figure.figsize'] = (10, 6)

🎯 Ek Nazar Mein — Poore Playlist ka Nichod¶

NumPy (kya seekha)¶

Array banane ke 8 tareeqe, attributes, dtype se memory optimize karna
Vectorized operations — loops ki zaroorat nahi
Boolean indexing + fancy indexing se filtering
Broadcasting ke 3 rules
ML formulas (sigmoid, MSE, cross-entropy) NumPy mein likhna
30+ utility functions (where, argmax, cumsum, percentile, set functions...)

Pandas (kya seekha)¶

Series & DataFrame — do core objects
loc vs iloc ka farq
Boolean masking se filtering
Missing values handle karna (dropna, fillna)
GroupBy — split/apply/combine
Merge & concat — tables jodna
MultiIndex, melt, pivot_table — data reshape karna
.str aur .dt accessors

Matplotlib (kya seekha)¶

6 basic plots: line, scatter, bar, hist, pie, box
Subplots aur object-oriented API
3D plotting aur contour
Customization — colors, styles, annotations

Seaborn (kya seekha)¶

Figure-level vs Axes-level ka farq
5 plot categories: relational, distribution, categorical, regression, matrix
hue, col, row se multi-dimensional analysis
pairplot + heatmap(corr) — EDA ke do sabse powerful tools

🔥 Interview ke liye Top Questions¶

NumPy: 1. NumPy list se fast kyun hai? (C implementation, contiguous memory, vectorization, SIMD) 2. Broadcasting ke rules kya hain? 3. ravel() vs flatten() — view vs copy 4. np.where() kaise kaam karta hai? 5. axis=0 aur axis=1 mein kya farq hai? 6. View aur copy mein kya antar hai?

Pandas: 1. loc vs iloc — label vs position, aur slicing mein end include/exclude 2. merge vs join vs concat — kab kya use karein 3. apply vs map vs applymap 4. pivot vs pivot_table ka farq 5. Missing values handle karne ke tareeqe 6. groupby().agg() mein multiple aggregations kaise 7. SettingWithCopyWarning kyun aata hai? 8. Memory optimize kaise karein? (astype, category dtype)

Visualization: 1. Histogram vs Bar chart mein kya farq hai? 2. Box plot kaise padhte hain? Outlier kaise identify karein? 3. Seaborn mein figure-level aur axes-level functions ka farq 4. Correlation heatmap se kya insight milta hai? 5. Kaunsa plot kab use karein? (num vs num, num vs cat, cat vs cat)