Metadata-Version: 2.1
Name: benchmarkit
Version: 0.0.4
Summary: Benchmark and analyze functions' time execution and results over the course of development
Home-page: https://github.com/vgrabovets/benchmarkit
Author: Vitaliy Grabovets
Author-email: v.grabovets@gmail.com
License: MIT
Keywords: benchmark,timeit,time
Platform: UNKNOWN
Requires-Python: >=3.6.0
Description-Content-Type: text/markdown
Requires-Dist: colorama (==0.4.1)
Requires-Dist: GitPython (==2.1.15)
Requires-Dist: pandas (==0.24.2)
Requires-Dist: path.py (==12.0.1)

# Benchmarkit

[![PyPI version](https://badge.fury.io/py/benchmarkit.svg)](https://badge.fury.io/py/benchmarkit)
[![Build Status](https://travis-ci.org/vgrabovets/benchmarkit.svg?branch=master)](https://travis-ci.org/vgrabovets/benchmarkit)
[![codecov](https://codecov.io/gh/vgrabovets/benchmarkit/branch/master/graph/badge.svg)](https://codecov.io/gh/vgrabovets/benchmarkit)
[![CodeFactor](https://www.codefactor.io/repository/github/vgrabovets/benchmarkit/badge)](https://www.codefactor.io/repository/github/vgrabovets/benchmarkit)
[![Dependabot Status](https://api.dependabot.com/badges/status?host=github&repo=vgrabovets/benchmarkit)](https://dependabot.com)

Benchmark and analyze functions' time execution and results over the course of development. 

## Features

- No boilerplate code
- Saves history and additional info
- Saves function output and parameters to benchmark data science tasks
- Easy to analyze results
- Disables garbage collector during benchmarking

## Motivation

- I need to benchmark execution time of my function
- I don't want to memorize and write boilerplate code
- I want to compare results with previous runs before some changes were introduced
- I don't want to manually write down results somewhere
- I want to know exact commits of my previous runs months ago
- I want to benchmark accuracy, precision, recall of my models and keep track of hyperparameters 

## Installation

```text
pip install benchmarkit
```

## Usage
### Benchmark execution times

Put `@benchmark` decorator over function with piece of code that should be timed

```python
from benchmarkit import benchmark, benchmark_run

N = 10000
seq_list = list(range(N))
seq_set = set(range(N))

SAVE_PATH = '/tmp/benchmark_time.jsonl'


@benchmark(num_iters=100, save_params=True, save_output=False)
def search_in_list(num_items=N):
    return num_items - 1 in seq_list


@benchmark(num_iters=100, save_params=True, save_output=False)
def search_in_set(num_items=N):
    return num_items - 1 in seq_set
```

- __num_iters__ - how many times to repeat benchmarked function. Default _1_
- __save_params__ - save parameters passed to the benchmarked function in the file with benchmark results. In the example above `num_items` will be saved. Default _False_
- __save_output__ - save benchmarked function output. Should return dict `{'name': value}`. Default _False_. See example how to benchmark model results.

Run benchmark:

```python
benchmark_results = benchmark_run(
    [search_in_list, search_in_set],
    SAVE_PATH,
    comment='initial benchmark search',
    rows_limit=10,
    extra_fields=['num_items'],
    metric='mean_time',
    bigger_is_better=False,
)  
```

- __functions__ - function or list of functions with `benchmark` decorator
- __save_file__ - path to file where to save results
- __comment__ - comment to save alongside the results
- __rows_limit__ - limit table rows in console output. Default _10_
- __extra_fields__ - extra fields to include in console output
- __metric__ - metric which is used for comparison. Default `mean_time`
- __bigger_is_better__ - whether bigger value of metric indicates that result is better. For time benchmarks should be `False`, for model accuracy should be `True`. Default _False_

Prints to terminal and returns list of dictionaries with data for the last run.

![Benchmark time output1](https://raw.githubusercontent.com/vgrabovets/benchmarkit/master/img/benchmark_time1.jpg)

Change `N=1000000` and rerun

![Benchmark time output2](https://raw.githubusercontent.com/vgrabovets/benchmarkit/master/img/benchmark_time2.jpg)

The same can be run from command line:
```text
benchmark_run test_data/time/benchmark_functions.py --save_dir /tmp/ --comment "million items" --extra_fields num_items
```

### Benchmark model results

```python
from benchmarkit import benchmark, benchmark_run
from sklearn.datasets import load_iris
from sklearn.linear_model import LogisticRegression

MODEL_BENCHMARK_SAVE_FILE = '/tmp/benchmark_model.jsonl'

x, y = load_iris(return_X_y=True)

@benchmark(save_params=True, save_output=True)
def log_regression(C=1.0, fit_intercept=True):
    clf = LogisticRegression(
        random_state=0, 
        solver='lbfgs', 
        multi_class='multinomial', 
        C=C,
        fit_intercept=fit_intercept,
    )
    clf.fit(x, y)
    score = clf.score(x, y)
    return {'score': score}

model_benchmark_results = benchmark_run(
    log_regression,
    MODEL_BENCHMARK_SAVE_FILE,
    comment='baseline model',
    extra_fields=['C', 'fit_intercept'],
    metric='score',
    bigger_is_better=True,
)
```

![Benchmark model1](https://raw.githubusercontent.com/vgrabovets/benchmarkit/master/img/benchmark_model1.jpg)

Change hyperparameter `C=0.5` and rerun. Output:

![Benchmark model2](https://raw.githubusercontent.com/vgrabovets/benchmarkit/master/img/benchmark_model2.jpg)

The same can be run from command line:
```text
benchmark_run file_with_benchmark.py --save_dir /tmp/ --comment "stronger regularization" --extra_fields C fit_intercept --metric score --bigger_is_better
```

### Analyze results from the file

```python
from benchmarkit import benchmark_analyze

SAVE_PATH = '/tmp/benchmark_time.jsonl'

benchmark_df = benchmark_analyze(
    SAVE_PATH,
    func_name=None, 
    rows_limit=10,
    metric='mean_time',
    bigger_is_better=False,
    extra_fields=['num_items'],
)
```

- __input_path__ - path to `.jsonl` file or directory with `.jsonl` files with benchmark results 
- __func_name__ - display statistics for particular function. If `None` then all functions, stored in file, are displayed. Default _None_
- __rows_limit__ - limit table rows in console output. Default _10_
- __metric__ - metric which is used for comparison. Default `mean_time`
- __bigger_is_better__ - whether bigger value of metric indicates that result is better. For time benchmarks should be `False`, for model accuracy should be `True`. Default _False_
- __extra_fields__ - extra fields to include in console output

Prints to terminal and returns pandas `DataFrame`.

![Benchmark analyze](https://raw.githubusercontent.com/vgrabovets/benchmarkit/master/img/benchmark_analyze.jpg)

The same can be run from command line:
```text
benchmark_analyze /tmp/benchmark_time.jsonl --extra_fields num_items
```

[Other examples](https://nbviewer.jupyter.org/github/vgrabovets/benchmarkit/blob/master/notebooks/benchmark_examples.ipynb)


