Filtering, Selecting and Slicing Your Data Like a Ninja

Written by: Marlon Colca
Posted on 13 May 2025 - 3 months ago
python pandas analytics

Master the art of selecting rows and columns using Pandas. Learn `loc`, `iloc`, conditions, and slicing techniques.

Filtering, Selecting and Slicing Your Data Like a Ninja 🥷

Once your data is loaded and clean, the next step is to zoom in on what matters — rows, columns, and specific conditions.

In this post, we’ll cover:

✅ Selecting specific columns
✅ Filtering rows with conditions
✅ Using .loc[] and .iloc[]
✅ Combining multiple filters
✅ Slicing data like a pro

Let’s slice and dice!

---

📦 Load the dataset

import pandas as pd

df = pd.read_csv("prices_sample.csv")

---

🎯 Selecting columns

Just want one column?

df["price"]

Multiple columns?

df[["product_name", "price"]]

---

🧽 Filtering rows by condition

Let’s say you want all products that cost more than 5:

df[df["price"] > 5]

Now products that are in stock:

df[df["in_stock"] == True]

Or products not in stock and price below 4:

df[(df["in_stock"] == False) & (df["price"] < 4)]

---

🧠 Using .loc[] and .iloc[]

.loc[] is label-based.
.iloc[] is index-based.

Example with .loc[]:

# First row, all columns
df.loc[0]

# First 3 rows, specific columns
df.loc[0:2, ["product_name", "price"]]

Example with .iloc[]:

# First row
df.iloc[0]

# First 5 rows and first 3 columns
df.iloc[0:5, 0:3]

---

🪓 Slicing like a pro

Let’s say you want the top 10 most expensive products:

df.sort_values("price", ascending=False).head(10)

Or the 5 cheapest in stock:

df[df["in_stock"] == True].sort_values("price").head(5)

---

🧠 Pro tip: Use .query() for readable filters

You can also use .query() to write conditions in a cleaner way:

df.query("in_stock == True and price < 5")

Much easier to read and chain when things get complex.

---

✅ Summary

Filtering and selecting is one of the most common things you’ll do in Pandas.

• Use [], .loc[], .iloc[] to select rows and columns
• Use conditions to filter rows
• Use .query() for clarity in complex logic
• Combine with .sort_values() and .head() to get top/bottom lists

---

📌 What’s next?

In Part 5, we’ll group and summarize data:
➡️ averages per category, counts per brand, pivot tables and more.
Don’t miss it — things get real 🔥

---

🔜 Coming up next