Tensorflow vs. PyTorch: Research Trends

Interesting new findings/developments. I’m leaving this here without personal comment, but rather for others to read if they’re interested.

The attached quote is from the blog’s author, Director of Machine Learning at Neural Magic, Mark Kurtz: "It’s shocking to see just how far TensorFlow has fallen. The 2022 state of competitive machine learning report came out recently and paints a very grim picture – only 4% of winning projects are built with TensorFlow. This starkly contrasts with a few years ago, when TensorFlow owned the deep learning landscape. Why did this happen, though?

Overall, poor architectural decisions led to abandonment from the community, and a monopoly-style view of ML led to a further lack of adoption from necessary tool chains in the ML ecosystem. The TensorFlow team tried to fix all of this with the TensorFlow v2 refactor, but it was too little, too late, and it abandoned the core piece TensorFlow was still holding on to — legacy systems.

I explore this further in my latest blog post, so check it out and let me know your thoughts!"

Again, left without comment.

This is in no way to say I fully agree, or disagree with using Pytorch vs. Tensorflow.

If I’m to say anything, its: go with the framework that you feel offers you the best success and performance, and continually/successfully delivers it!