You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
PyTorch has a lot of marketing firepower behind it, and as a result there's a common misconception that it has "momentum". Does it? I can't tell for sure, but the handful of traction indicators I monitor are showing that its user base has likely peaked around April-May 2018https://twitter.com/vincenzomanzoni/status/1047561148990730246 …
This graph is PyTorch-related search queries, for instance. In general, all indicators are heavily correlated.pic.twitter.com/l9bUkRlArz
IMO @jeremyphoward's fastai library will be the next Keras and transitively popularize PyTorch
Why not be agnostic and have it support more than 1 framework?
Here is the same graph for TensorFlow. Pytorch is gaining momentum relative to TensorFlow, not necessarily in absolute numbers. Especially in the research community.pic.twitter.com/CILuDPF1jU
Both graphs are flat. However there is no ongoing narrative that "TF is growing fast"... TF has already reached market saturation (and the market has stopped growing).
I'm deeply confused about how you could possibly conclude the use base peaked from this data. Time series forecasting is like ... hard.
Google search queries over the same period of time for both frameworks:pic.twitter.com/GirjgUTqZ8
How do you explain the "heartbeat" pattern there? No-one looks for PyTorch on the weekend? ha ha
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
