Why do you think there are so many issues with Tensorflow ? (if you think so.)

Been following many issues. The threads go for years and people just randomly trying things.

I'm aware that we can just switch to other framework, but is there a simple reason why this framework has so many issues?

Once you get it working, it does most jobs very fast and reliably imho. But getting there is hard.

deeplearning framework people reason simple tensorflow things think threads you