So true for work and side-projects as well.
(I needed this reminder myself as I have a blog post I've been noodling on for 12+ months. I just need to publish the damned thing!)
So, if it weren't mandatory, handing it in isn't all that important as long as you're doing it. Indeed, in grad school, there was very little graded homework and most was just assigned with solutions given for anyone who wanted to do it.
So, to bring the overly stretched analogy back to projects, I think you shouldn't feel like you need to "finish" something if it meets enough of the goals you set for yourself.
Eg I spent months working on projects like custom OS's and game engines. But as the nature of those kinds of projects is to have endless room for growth, I just drew the line at some point that I had learned and done enough and could just drop them in whatever potentially buggy state they were in.
I actually have a project on the shelf that's ][ far from being releasable.
It's waiting on two things.
One, is since it's a Java GUI, getting it built into a form for the MacOS/Window/Linux platforms, rather than just "here's a jar file". There's ways around this, but it's not quite drag and drop, and I haven't got a Windows box to test anything on, so its stalled. (I tried installing one of MS's VM images that they offered, but that's didn't work, so...back to the shelf).
Two, is documentation. I could probably just let it go, and explain it to the 3 people who might actually download it and use because simply launching it, "Neat!", and letting it bit rot forever more. But, I think it needs some documentation, so...it sits.
If I could satisfactorily drag and drop the installers, I'd probably press on with it, but it could still use some docs.
Meanwhile, I continue on my meta project which this project would inevitably be folded into, so it's not all lost.
It is really not a good experience, but totally beats not having the software.