How can I use tracemalloc for memory profiling in Python?

How can I use tracemalloc for memory profiling in Python?

You can use the tracemalloc module to trace memory allocations in your Python code. First, import it and start tracing with tracemalloc.start(). Then, after your code runs, you can use tracemalloc.get_traced_memory() to get the current and peak memory usage.

Can you show me a simple example of how to implement it?

Sure! Here's a simple example:

import tracemalloc

tracemalloc.start()

# Your code here

snapshot = tracemalloc.take_snapshot()
for stat in snapshot.statistics('lineno'):
    print(stat)
``` This will print memory usage statistics by line number.

What if I want to compare memory usage between two different code snippets?

You can take two snapshots at different points in your code. For example:

tracemalloc.start()
# First code snippet
snapshot1 = tracemalloc.take_snapshot()
# Second code snippet
snapshot2 = tracemalloc.take_snapshot()

stats = snapshot2.compare_to(snapshot1, 'lineno')
for stat in stats:
    print(stat)
``` This will show you the differences in memory usage.

Are there any limitations I should be aware of?

Yes, tracemalloc can add some overhead to your program's performance, especially if you're tracing a lot of allocations. It's best used for debugging and profiling rather than in production.