NumPy Boolean Slicing – Filter Arrays Using Conditions in Python

🐍 NumPy – Slicing with Boolean Arrays

Boolean slicing in NumPy is a powerful technique used to filter arrays based on conditions.

Instead of using loops, NumPy allows you to apply conditions directly to arrays and extract only the required elements.

This technique is also known as:

• Boolean indexing
• Masking
• Conditional filtering

It is widely used in:

• Data science
• Machine learning
• Data cleaning
• Image processing
• Statistical analysis

---

What is Boolean Slicing?

Boolean slicing means using a boolean array (True/False values) to select elements from a NumPy array.

• True → element is selected
• False → element is ignored

---

Example Concept

Array:   [10, 20, 30, 40, 50]
Condition: > 25

Result:  [30, 40, 50]

---

🔵 Creating Boolean Arrays

A boolean array is created by applying a condition.

Example

import numpy as np

arr = np.array([10, 20, 30, 40, 50])

print(arr > 25)

Output:

[False False  True  True  True]

---

🟢 Using Boolean Array for Filtering

Now use this boolean array to filter values.

result = arr[arr > 25]

print(result)

Output:

[30 40 50]

---

🟡 Multiple Conditions (AND)

Use & operator for AND condition.

arr = np.array([10, 15, 20, 25, 30, 35, 40])

result = arr[(arr > 15) & (arr < 40)]

print(result)

Output:

[20 25 30 35]

---

🔴 Multiple Conditions (OR)

Use | operator for OR condition.

result = arr[(arr < 15) | (arr > 35)]

print(result)

Output:

[10 40]

---

⚠️ Important Rule: Use Parentheses

When combining conditions, always use parentheses.

(arr > 10) & (arr < 40)

✔ Correct
❌ arr > 10 & arr < 40 (Wrong)

---

🟣 Boolean Slicing in 2D Arrays

arr = np.array([
    [10, 20, 30],
    [40, 50, 60],
    [70, 80, 90]
])

print(arr[arr > 50])

Output:

[60 70 80 90]

---

🟤 Replace Values Using Boolean Mask

You can also modify data using conditions.

arr = np.array([10, 20, 30, 40, 50])

arr[arr > 30] = 99

print(arr)

Output:

[10 20 30 99 99]

---

🔵 Boolean Masking for Data Cleaning

data = np.array([10, -5, 20, -1, 30])

cleaned = data[data > 0]

print(cleaned)

Output:

[10 20 30]

✔ Removes invalid or negative values

---

🧠 How Boolean Indexing Works

Step-by-step process:

1. Apply condition on array
2. NumPy creates boolean mask
3. Mask is applied to filter data
4. Only True values are returned

---

📊 Boolean Slicing vs Fancy Indexing

Feature	Boolean Slicing	Fancy Indexing
Based on	Condition	Index list
Selection	Dynamic	Manual
Flexibility	High	High
Use case	Filtering	Specific selection

---

⚡ Real-World Example

Filtering student marks

marks = np.array([45, 67, 89, 30, 55, 90])

passed = marks[marks >= 50]

print(passed)

Output:

[67 89 55 90]

---

🖼️ Image Processing Example

image = np.array([
    [100, 150, 200],
    [50,  80,  120],
    [30,  60,  90]
])

bright = image[image > 100]

print(bright)

Output:

[150 200 120]

✔ Used in computer vision for pixel filtering

---

🚀 Performance Tips

• Boolean slicing is vectorized (very fast)
• Avoid loops when filtering data
• Use .copy() if you need independent data
• Combine conditions carefully for efficiency

---

⚠️ Common Errors

1. Missing parentheses

ValueError or unexpected result

✔ Fix:

(arr > 10) & (arr < 20)

---

2. Using Python keywords instead of operators

❌ and, or
✔ Use &, |

---

🧾 Summary

NumPy boolean slicing allows you to:

• Filter arrays using conditions
• Apply logical operations
• Clean and transform data
• Modify values efficiently

---

🏁 Conclusion

Boolean slicing is one of the most powerful features in NumPy. It enables fast, readable, and efficient data filtering without loops.

It is essential for:

• Data science
• Machine learning
• Data preprocessing
• Image processing