tprofiler 1.0.3

A combined time and memory profiler using psutil

tprofiler

tprofiler is a lightweight Python library for total profiling—combining time and memory profiling using psutil with optional line-by-line profiling using line_profiler. It provides a decorator for profiling individual functions, a context manager for profiling code blocks, and a command-line tool for profiling entire scripts.

Features

• Combined Time and Memory Profiling: Track execution time and process memory (RSS) before and after function or code block execution.
• Easy-to-Use Decorator: Simply add @profile to any function to get detailed profiling output.
• Line-by-Line Profiling: Use @profile.line to obtain a detailed, line-by-line performance analysis (requires line_profiler).
• Context Manager: Profile arbitrary code blocks with the provided ProfileContext.
• Command-Line Tool: Run any Python script with tprofiler to obtain an overall profiling summary.

Installation

tprofiler is available on PyPI. Install it using pip:

pip install tprofiler

Usage

1. As a Decorator

Add profiling to any function by importing and applying the decorator:

from tprofiler.core import profile

@profile(enable_memory=True, enable_time=True, verbose=True)
def my_function(n):
    total = sum(range(n))
    return total

result = my_function(1000000)

When my_function is called, tprofiler prints the execution time and memory usage details, along with the function's return value if verbose is enabled.

2. Line-by-Line Profiling

For a detailed line-by-line analysis, use the line profiling decorator:

from tprofiler.core import profile

@profile.line
def compute_heavy(n):
    data = [i for i in range(n)]
    return sum(data)

compute_heavy(10_000_000)

This will output detailed line-by-line performance statistics for compute_heavy (ensure line_profiler is installed).

3. As a Command-Line Tool

You can profile an entire script by running:

tprofiler your_script.py [script arguments...]

For example, if you have a script named example.py, run:

tprofiler example.py --option value

This command executes the script and prints an overall profiling summary including total time elapsed and memory consumption.

4. Using the Context Manager

To profile a block of code without decorating a function, use the ProfileContext:

from tprofiler import ProfileContext

with ProfileContext(enable_memory=True, enable_time=True):
    # Place the code you want to profile here
    total = sum(range(1000000))
    print(total)

How It Works

• Time Profiling:

  Uses Python's time module to capture the execution time before and after function calls or code blocks.

• Memory Profiling:

  Uses psutil to measure the process's memory usage (RSS) before and after execution.

• Line Profiling:

  Integrates with line_profiler to provide detailed per-line execution statistics.

Contributing

Contributions and improvements are welcome! Feel free to open issues or submit pull requests on GitHub.

License

This project is licensed under the MIT License.