3 ms·
Both PyTorch and Tensorflow are used extensively. If you look at the last year's most popular models - GPT2 and StyleGan - they are both Tensorflow. When I chec
by Tenoke 7y ago
Both PyTorch and Tensorflow are used extensively. If you look at the last year's most popular models - GPT2 and StyleGan - they are both Tensorflow. When I check on Scholar there are many, many results for both Tensorflow and PyTorch from 2019. The most recent stuff I've been working on (because that's what they were published in) have been in Jax.
PyTorch is one of the biggest frameworks but it's unreasonable to suggest it is the only one that matters.
- _coveredInBees 7y agoGPT-2 was developed by Open-AI. They recently switched to Pytorch themselves for similar reasons [1]. I didn't claim that all research that matters is happening on Pytorch, but it is very true that most non-Google research is [2]. Having used both, Pytorch has been a far better experience for me, and lately, most of the reference implementations I care about are released in Pytorch, so it makes it very easy to experiment with and also pull into my own code base if needed. Also, for an intro class, Pytorch would have been a better choice imo, because it has a simpler and cleaner API. In my experience, it was a lot easier to get up and running with Pytorch compared to Tensorflow. 1 - https://openai.com/blog/openai-pytorch/ https://openai.com/blog/openai-pytorch/ 2 - https://news.ycombinator.com/item?id=21216200 https://news.ycombinator.com/item?id=21216200