📊 Understanding pandas: pd, DataFrames, and Methods (Beginner-Friendly Guide)

If you’ve ever written code like:

df.describe()["age"]

and wondered:

“Why is describe() called on df and not pd?”
“Are DataFrames not part of pandas (pd)?”

This post will clear that confusion completely.

---

🔹 Step 1: What is pd?

When you write:

import pandas as pd

👉 pd is just an alias for the pandas library.

Think of pd as a toolbox 🧰 that contains many tools and structures.

---

🔹 Step 2: What is a DataFrame?

Inside pandas, there is a class called:

pd.DataFrame

This is a blueprint for creating tables.

When you do:

df = pd.DataFrame({
    "age": [25, 30, 35],
    "salary": [40000, 50000, 60000]
})

👉 df becomes an actual DataFrame object

So:

• pd.DataFrame → blueprint
• df → real object created from that blueprint

✔ Yes — DataFrames are absolutely part of pandas.

---

🔹 Step 3: Two types of functions in pandas

This is where most confusion happens.

✅ 1. Functions of pd (module-level)

These are general tools:

pd.read_csv()
pd.concat()
pd.to_datetime()

👉 They are not tied to a specific DataFrame
👉 They help you create or manipulate data

---

✅ 2. Methods of a DataFrame

Once you have a DataFrame (df), it comes with its own methods:

df.describe()
df.head()
df.mean()

👉 These belong to the DataFrame object itself

---

🔹 Step 4: Why df.describe() and NOT pd.describe()?

Because:

• describe() is defined inside the DataFrame class
• It operates on a specific dataset (df)

❌ Wrong:

pd.describe()

✔ Correct:

df.describe()

---

🔹 Step 5: Understanding df.describe()["age"]

Let’s break it step-by-step:

df.describe()

👉 Returns a new DataFrame (summary statistics)

	age	salary
mean	30	50000
std	…	…

Then:

["age"]

👉 Selects the "age" column from that result

So:

df.describe()["age"]

means:

“Generate a summary → then extract only the age column”

---

🔹 Step 6: Better alternative

Instead of:

df.describe()["age"]

You can write:

df["age"].describe()

👉 This directly describes only one column (cleaner approach)

---

🔹 Step 7: Simple analogy (this will stick)

Think of:

• pd = 🏭 Factory
• DataFrame = 🚗 Car design
• df = 🚘 Your car

Then:

• pd.read_csv() → factory tool
• df.describe() → your car’s feature

---

🔹 Final Takeaway

• pd = the library (toolbox)
• DataFrame = structure inside pandas
• df = actual data object
• df.describe() = method acting on your data
• ["age"] = selecting from the result

---

🚀 Key Insight

👉 () → used to call methods/functions
👉 [] → used to select data from objects

---