Any experience (good or bad) using TF for unsupervised learning?

Having used many unsupervised learning algorithms in the past as part of my development pipeline, I was wondering about weather there are good implementations out there that are built on top of TensorFlow (I’ve seen a couple of k-means implementations in tutorial format, but not much more).

What tools other than TensorFlow do you use for your unsupervised learning needs?


I have been using TensorFlow for coding up a number of different self-supervised models for vision and the experience have been great (and easy). Feel free to take a look into the following minimal implementations of popular self-supervised methods for vision:

Sorry, could not post links because the forum does not yet allow it.