Ask HN: Is PyTorch better than TensorFlow for production?

Re:Ask HN: Is PyTorch better than TensorFlow for production? by SomeoneFromCA on Friday December 04, 2020 @02:43PM (#25306775)

I used Darknet (C library) and good old C++ and CUDA custom written NNs. I've never had any success with Python frameworks, they all felt fragile and not very performant.

[ Parent | on HN ]

Re:Ask HN: Is PyTorch better than TensorFlow for production? by Jugurtha on Saturday December 05, 2020 @12:16AM (#25311987)

How frequently do you use that, and do you do so for projects with stakes? As in, you do it professionally where results are expected. If so, you probably work within a team. Does your team also use this to do their work? How do you ship things?

[ Parent | on HN ]

Re:Ask HN: Is PyTorch better than TensorFlow for production? by sgillen on Friday December 04, 2020 @05:26PM (#25309100)

Interesting, have you tried torches C++ API at all? I’m wondering if the frameworks are fragile or just python the language .

[ Parent | on HN ]

Re:Ask HN: Is PyTorch better than TensorFlow for production? by SomeoneFromCA on Saturday December 05, 2020 @07:53AM (#25314177)

It was a nightmare to use from C++, and yeah, it was fragile. Darknet is very simple, fast and get things done, although it doesnt feel like high quality product either.

[ Parent | on HN ]