14 engineers on Hacker News describe using TensorFlow first-hand: 1 endorse it, 4 regret it, 1 are mixed.
“I've done a few work projects in Keras/TensorFlow over the years and it created a lot of technical debt and lost time debugging it”
regret · production · minimaxir · 2024-11-13 · source ↗ · wrong label?“Problem with TensorFlow is it is mostly not readable and onboarding new people to project is really hard.”
regret · production · sercand · 2020-01-30 · source ↗ · wrong label?“Back when we were using TensorFlow, whenever we wanted to try something new... we would find ourselves wrestling with its machinery”
regret · production · cs702 · 2020-01-30 · source ↗ · wrong label?“we would find ourselves wrestling with its computational graph abstraction, which is non-intuitive, especially for models with more complex control flow.”
regret · production · cs702 · 2019-10-10 · source ↗ · wrong label?“if you are serious about getting into data science or AI I have to agree with the OP -- stick to TensorFlow.”
endorse · production · bluejay2387 · 2018-12-28 · source ↗ · wrong label?“At work, we have a bunch of vectorized computations that we run in Tensorflow, as it's a convenient way to get GPU-optimized code, so that is still an option (albeit an awkward one).”
mixed · production · fnbr · 2018-06-28 · source ↗ · wrong label?“debug a tensorflow TPU problem with an oscilloscope connected to the voltage regulator in a hardware lab”
neutral · production · dekhn · 2022-08-04 · source ↗ · wrong label?“These are large TensorFlow models... custom compiled TensorFlow container images.”
neutral · production · ra-mos · 2021-12-27 · source ↗ · wrong label?“all the research papers are accompanied by complete tensorflow/pytorch implementations so we only had to do minor modifications to train/test them on our dataset.”
neutral · personal · krut-patel · 2019-12-25 · source ↗ · wrong label?“Models are written in Python (mix of pytorch/NLP/tensorF).”
neutral · production · superkitty · 2019-11-12 · source ↗ · wrong label?“We don't take too strong an opinion on ML framework and use both Tensorflow and Pytorch currently.”
neutral · production · ayw · 2019-08-05 · source ↗ · wrong label?“We have released with tensorflow, we will be releasing with Keras in the coming week”
neutral · production · rahults · 2017-12-01 · source ↗ · wrong label?“we use tools like Tensorflow and Keras, similar to most AI research labs.”
neutral · production · brandonb · 2016-12-01 · source ↗ · wrong label?