# Tensorflow vs. PyTorch

**URL:** <https://ask.cyberinfrastructure.org/t/tensorflow-vs-pytorch/1190>\
**Category:** Q&A\
**Tags:** machine-learning, tensorflow, researcher\
**Created:** [December 11, 2019, 9:19pm UTC](https://ask.cyberinfrastructure.org/t/tensorflow-vs-pytorch/1190 "2019-12-11T21:19:49Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![jb4earth](https://ask.cyberinfrastructure.org/user_avatar/ask.cyberinfrastructure.org/jb4earth/32/378_2.png) [@jb4earth](https://ask.cyberinfrastructure.org/u/jb4earth)\
**Post date:** [December 11, 2019, 9:19pm UTC](https://ask.cyberinfrastructure.org/t/tensorflow-vs-pytorch/1190/1 "2019-12-11T21:19:49Z")

</div>

What are you using and why?

---

<div class="post-metadata">

**Author:** ![kir0ul](https://ask.cyberinfrastructure.org/user_avatar/ask.cyberinfrastructure.org/kir0ul/32/380_2.png) [@kir0ul](https://ask.cyberinfrastructure.org/u/kir0ul)\
**Post date:** [December 12, 2019, 3:31pm UTC](https://ask.cyberinfrastructure.org/t/tensorflow-vs-pytorch/1190/2 "2019-12-12T15:31:02Z")

</div>

If that might help: [https://thegradient.pub/state-of-ml-frameworks-2019-pytorch-dominates-research-tensorflow-dominates-industry/](https://thegradient.pub/state-of-ml-frameworks-2019-pytorch-dominates-research-tensorflow-dominates-industry/)

---

<div class="post-metadata">

**Author:** ![jb4earth](https://ask.cyberinfrastructure.org/user_avatar/ask.cyberinfrastructure.org/jb4earth/32/378_2.png) [@jb4earth](https://ask.cyberinfrastructure.org/u/jb4earth)\
**Post date:** [December 12, 2019, 11:14pm UTC](https://ask.cyberinfrastructure.org/t/tensorflow-vs-pytorch/1190/3 "2019-12-12T23:14:55Z")

</div>

Great article. Thank you for sharing. I knew academia was more into pytorch, but this gives some real proof.

Throwing Jax in at the end there really through me for a loop! I hadn’t even thought of higher-order gradient schemes or pushing numpy maths onto the GPU. Will definitely be using a few functions from that library.