checkpointe

checkpointe

A simple progress timer/memory tracker for debugging python scripts.

Overview

Checkpointe is a tool for locating areas of code that are time or memory-intensive. This allows the user to:

• Track specific function calls in real-time
• Tag specific checkpoints with an identifying message
• Analyze a final output that shows time and percentage of time between each checkpoint
• Reorder checkpoints easily without having to refactor code
• Analyze memory usage to determine where usage is peaking

Requires

• datetime
• tracemalloc (for memory tracking)

Installation

Checkpointe is available for download using pip:

pip install checkpointe

Using Checkpointe

Start

To start tracking time and/or memory, initialize checkpointe with the .start() function

import checkpointe as check

check.start(summary=True, verbose=True, memory=True)

Setting these defaults determines the level of detail checkpointe prints to stdout:

• summary (default = True): prints a summary after calling the .stop() method
• verbose (default = False): if enabled, outputs data at each check point
• memory (default = False): if enabled, tracks and outputs information about system memory usage

Point

Call the .point() method to create a marker at any point in the code.

add = 2 + 2

check.point("ADDITION")

mult = 2 * 2

check.point("MULTIPLICATION")

exp = 2 ** 2

check.point("EXPONENTIAL")

If verbose=True, a statement will be printed at each checkpoint.

Stop

When you are ready to stop tracking, call the .stop() method.

check.stop()

If summary=True, a summary statement will be printed. Each marker will be displayed, along with the time elapsed since the previous marker and the percentage of overall time elapsed.

If memory=True, a final memory statement will be printed, with memory usage at each marker, along with the peak memory usage noted at each marker.

Future Development

Future versions will enable integration with logging to an output file.