remove repos with less stars

[zhimin-z]

No description provided.

[axsaucedo]

I am also considering reverting this PR as it was discussed and it was decided not to go forward.

However looking at the list of removals I can see that it's a pretty reasonable reduction.

Let me have a think, as I think it's a reasonable change, but I somehow want to also ensure the bar is not to high for "high potential projects that are getting started", as lists like these can also serve as a catapult to increase engagement with some libraries

[zhimin-z]

I am also considering reverting this PR as it was discussed and it was decided not to go forward.

However looking at the list of removals I can see that it's a pretty reasonable reduction.

Let me have a think, as I think it's a reasonable change, but I somehow want to also ensure the bar is not to high for "high potential projects that are getting started", as lists like these can also serve as a catapult to increase engagement with some libraries

Hi, @axsaucedo Thanks for your feedback. I took a look but found some tools in our previous list are less impactful (<300 stars), and they are not actively maintained (last update >4 months) as well. Although they have existed for many years (creation date >5 years) but is not as prevalent as the remaining ones since nobody really use/adapt it (< 100 PRs, <10 contributors).

Thus, I propose that we might need to retrospect our standard of "production-level" since not every tool that have 100 stars are worth in our list (considering our current list is already too long and verbose). In other words, if a tool is production-ready, but no one likes to use it in production environment, then it is high time to not consider them as production-level ML tools in this list.

[axsaucedo]

I took a look but found some tools in our previous list are less impactful (<300 stars)
Yeah looking at the PR I don't think it would add a significant issue.

In general my main perspective is whether there are exceptions where adding a library with >300<500 stars can also support the library as lists like these tend to be used for high potential libraries to also grow.

However I currently don't see a particular example of this, and we don't really get a lot of inbound libraries so we can re-asses if we do see a lot of exceptions cropping up.

For now I am happy to proceed on this - thanks @zhimin-z.