If you’ve ever wondered whether to use plain Python lists or NumPy arrays for numerical computations, you’re not alone. In this tutorial, we put them head-to-head in a series of speed tests. You’ll see exactly when NumPy becomes the clear winner, especially with large datasets.
🎓 What Are We Testing?
We’ll compare Python lists and NumPy arrays in three operations:
- Summing elements
- Element-wise multiplication
- Memory usage
Each test runs on a dataset with 10 million values to simulate real-world data loads like scientific computing or image processing.
⚙️ Setup and Data Generation
Import libraries and generate data
import time
import numpy as np
# Data size
N = 10_000_000
# Python list and NumPy array
py_list = list(range(N))
np_array = np.arange(N)We create equivalent datasets in both formats so the comparison is fair.
⏱️ Speed Test 1 — Summing Elements
Using Python list:
start = time.time()
sum_py = sum(py_list)
end = time.time()
print("Python list sum:", end - start, "seconds")Using NumPy:
start = time.time()
sum_np = np.sum(np_array)
end = time.time()
print("NumPy sum:", end - start, "seconds")NumPy significantly outperforms native Python thanks to low-level optimizations.
🔢 Speed Test 2 — Element-wise Multiplication
Using Python list:
start = time.time()
py_mult = [x * 2 for x in py_list]
end = time.time()
print("Python list multiplication:", end - start, "seconds")Using NumPy:
start = time.time()
np_mult = np_array * 2
end = time.time()
print("NumPy array multiplication:", end - start, "seconds")No loops, no hassle — NumPy performs these operations in compiled C, making it lightning fast.
📊 Memory Efficiency Comparison
Let’s compare how much memory each structure uses:
print("Python list size:", len(py_list) * py_list[0].__sizeof__(), "bytes")
print("NumPy array size:", np_array.nbytes, "bytes")NumPy arrays are not only faster but also more compact in memory, which is crucial for data-heavy tasks.
📄 Wrap-Up
NumPy isn’t always the right tool — but when dealing with large numerical arrays, it’s a game-changer. You get:
- Faster computation
- Cleaner, vectorized code
- Lower memory usage
If you’re working with math-heavy applications, NumPy is a must-learn.
📌 Summary
- Compared Python lists and NumPy arrays
- Benchmarked summing and multiplying operations
- Measured memory usage
- Observed significant performance gains with NumPy
🔗 Resources
- NumPy Documentation
- NumPy Tutorials (Real Python)
- Python Time Complexity Wiki
Stick around for our next deep dive into vectorized operations and broadcasting in NumPy!