Performance Optimization

Introduction

Techniques for optimizing Python application performance.

Profiling

import cProfile
import pstats

# Profile function
cProfile.run("my_function()", "stats.prof")

# Analyze results
pstats.Stats("stats.prof").sort_stats("cumulative").print_stats(10)

Line Profiler

# pip install line_profiler
# Add @profile decorator
from line_profiler import LineProfiler

profiler = LineProfiler()
profiler.add_function(my_function)
profiler.runcall(my_function, args)
profiler.print_stats()

Optimization Tips

# Local variable access is faster
def fast_function():
    local_var = expensive_operation
    for item in items:
        result = local_var + item  # Faster than global
    return result

# Use list comprehensions instead of loops
result = [x ** 2 for x in range(1000)]

# Use join instead of concatenation
parts = ["a", "b", "c"]
result = "".join(parts)

Practice Problems

1. Profile code to find bottlenecks
2. Optimize hot paths
3. Use appropriate data structures
4. Reduce function call overhead
5. Use compiled extensions