PyTorch and TensorFlow usage in 2019 #PyTorch #TensorFlow #MachineLearning #DataScience #ArtificialIntelligence @cHHillee
Horace He (@cHHillee) recently published an article on The Gradient that looks at trends of machine learning frameworks in academia vs industry. While TensorFlow dominates in industry, there are some striking trends suggesting rapid growth with the PyTorch framework in research. @cHHillee outlines stats for conferences that occurred last year to support this point:
In 2018, PyTorch was a minority. Now, it is an overwhelming majority, with 69% of CVPR using PyTorch, 75+% of both NAACL and ACL, and 50+% of ICLR and ICML.
While PyTorch is seeing success in research, TensorFlow still has higher usage overall (likely driven by industry) with a larger number of job listings, medium articles, and GitHub stars:
…from 2018 to 2019, TensorFlow had 1541 new job listings vs. 1437 job listings for PyTorch on public job boards, 3230 new TensorFlow Medium articles vs. 1200 PyTorch, 13.7k new GitHub stars for TensorFlow vs 7.2k for PyTorch, etc.
@cHHillee goes on to discuss some potential reasons for the preference of PyTorch over TensorFlow in research. The ease of use, simplicity, and performance of PyTorch are all outlined as potential contributors. However, a number of the challenges researchers face with the TensorFlow framework may be absolved with the release of TensorFlow 2.0. on that note, some of the weaknesses of PyTorch were fixed with the introduction of the JIT compiler and TorchScript. @cHHillee suggests that this exhibits convergence of the frameworks. As it stands, both frameworks have momentum in the ML community with strong support. It is likely industry/field-dependent which framework makes the most sense. If you would like to read more comparisons between TensorFlow and PyTorch check out these articles:
Stop breadboarding and soldering – start making immediately! Adafruit’s Circuit Playground is jam-packed with LEDs, sensors, buttons, alligator clip pads and more. Build projects with Circuit Playground in a few minutes with the drag-and-drop MakeCode programming site, learn computer science using the CS Discoveries class on code.org, jump into CircuitPython to learn Python and hardware together, TinyGO, or even use the Arduino IDE. Circuit Playground Express is the newest and best Circuit Playground board, with support for CircuitPython, MakeCode, and Arduino. It has a powerful processor, 10 NeoPixels, mini speaker, InfraRed receive and transmit, two buttons, a switch, 14 alligator clip pads, and lots of sensors: capacitive touch, IR proximity, temperature, light, motion and sound. A whole wide world of electronics and coding is waiting for you, and it fits in the palm of your hand.