speedopy 0.2.0

Zero-dependency performance measurement decorators for Python

speedopy

Zero-dependency performance measurement decorators for Python.

Just import a decorator, slap it on your function, and instantly see metrics printed to the console. No config, no setup, no external packages.

Install

pip install speedopy

Decorators

Decorator	What It Measures
cpu_time	CPU processing time (excludes I/O)
real_time	Wall-clock time
memory_usage	Peak memory allocated (auto-formats B/KB/MB)
thread_time	CPU time for the current thread only
call_count	Number of times a function has been called
function_profile	Full cProfile report (top 10 calls)
io_time	I/O wait time (real - cpu)
gc_stats	Garbage collections triggered
object_count	Net new Python objects created

Usage

from speedopy import cpu_time, memory_usage, io_time

@cpu_time
@memory_usage
@io_time
def process_data():
    data = [i ** 2 for i in range(1_000_000)]
    return sum(data)

process_data()

Output:

[io_time] process_data -> real: 0.182345s, cpu: 0.171875s, io_wait: 0.010470s
[memory_usage] process_data -> current: 472 B, peak: 8.06 MB
[cpu_time] process_data -> 0.183291s

Stack Multiple Decorators

from speedopy import cpu_time, memory_usage, call_count

@call_count
@cpu_time
@memory_usage
def my_func():
    return sum(range(500_000))

my_func()  # [call_count] my_func -> called 1 time(s)
my_func()  # [call_count] my_func -> called 2 time(s)

Requirements

• Python 3.8+
• No external dependencies

License

MIT