austin VS doit

I collected a list of profilers (also memory profilers, also specifically for Python) here: https://github.com/albertz/wiki/blob/master/profiling.md

Currently I actually need a Python memory profiler, because I want to figure out whether there is some memory leak in my application (PyTorch based training script), and where exactly (in this case, it's not a problem of GPU memory, but CPU memory).

I tried Scalene (https://github.com/plasma-umass/scalene), which seems to be powerful, but somehow the output it gives me is not useful at all? It doesn't really give me a flamegraph, or a list of the top lines with memory allocations, but instead it gives me a listing of all source code lines, and prints some (very sparse) information on each line. So I need to search through that listing now by hand to find the spots? Maybe I just don't know how to use it properly.

I tried Memray, but first ran into an issue (https://github.com/bloomberg/memray/issues/212), but after using some workaround, it worked now. I get a flamegraph out, but it doesn't really seem accurate? After a while, there don't seem to be any new memory allocations at all anymore, and I don't quite trust that this is correct.

There is also Austin (https://github.com/P403n1x87/austin), which I also wanted to try (have not yet).

Somehow this experience so far was very disappointing.

(Side node, I debugged some very strange memory allocation behavior of Python before, where all local variables were kept around after an exception, even though I made sure there is no reference anymore to the exception object, to the traceback, etc, and I even called frame.clear() for all frames to really clear it. It turns out, frame.f_locals will create another copy of all the local variables, and the exception object and all the locals in the other frame still stay alive until you access frame.f_locals again. At that point, it will sync the f_locals again with the real (fast) locals, and then it can finally free everything. It was quite annoying to find the source of this problem and to find workarounds for it. https://github.com/python/cpython/issues/113939)

Just wanted to add Austin: Python frame stack sampler for CPython written in pure C (https://github.com/P403n1x87/austin)

If you follow the call stack carefully you should be able to get to the point where sklearn calls ddot_kernel_8 (indirectly in this case). Austin(p) reports source files as well, so that shouldn't be a problem (provided all the debug symbols are available). If you're collecting data with austinp, don't forget to resolve symbol names with the resolve.py utility (https://github.com/P403n1x87/austin/blob/devel/utils/resolve..., see the README for more details: https://github.com/P403n1x87/austin/blob/devel/utils/resolve...)

I like mage for Go and doit for Python.

Some competitors - Rake (ruby) - Bake - Earthly - SCons - doit

An alternative to Scons could be Doit (<https://pydoit.org/>), which if I remember correctly was built as a faster alternative to Scons. See also reasons of some users to prefer the later to other mentioned here: <https://pydoit.org/stories.html>.

If you code in Python, your probably should use the language as much as possible and avoid calling shell commands.

E.G:

- manipulate the file system with pathlib

- do hashes with hashlib

- zip with zipfile

- set error code with sys.exit

- use os.environ for env vars

- print to stderr with print(..., file=...)

- sometimes you'll need to install lib. Like, if you want to manipulate a git repo, instead of calling the git command, use gitpython (https://gitpython.readthedocs.io/en/stable/)

But if you don't feel like installing a too many libs, or just really want to call commands because you know them well, then the "sh" lib is going to make things smoother:

https://pypi.org/project/sh/

Also, enjoy the fact Python comes with argparse to parse script arguments (or if you feel like installing stuff, use typer). It sucks to do it in bash .

If what you need is more build oriented, like something to replace "make", then I would instead recommend "doit":

https://pydoit.org/

It's the only task runner that I haven't run away from yet.

Remember to always to everything in a venv. But you can have a giant venv for all the scripts, and just she-bang the venv python executable so that it's transparent. Things don't have to be difficult.

I've been using Doit for a project which involves gathering together documents made up of multiple Markdown files and converting to multiple formats. It's really cool but has some irritations. It didn't end up being much simpler than Make for me. I'm interested in trying some of the alternatives people have posted.

The code for all of this is available here, and described in detail in my article. I'm particularly fan of doit for this type of project, and highly encourage everyone to check it out!

Instead, I use pydoit (which is basically a Python version of make). It's simple, flexible, and quite extensible. So, here's what I do with it: