Even if you don't care about machine learning, TensorFlow's XLA is amazing for farming code out to the GPU. GPGPU has never been easier.
TensorFlow 1.0 Released
61–70 of 76 posts
Re: TensorFlow 1.0 Released
#62Could MacBook Pros (with Intel HD Graphics 3000 384 MB, to be more specific) train with GPU? I've always wanted to train algorithms but without using the GPU it is really slow.
I doubt the integrated Intel Card would be supported, even if it is, using the CPU would be just as good if not better. A lot of the high performance you see on GPUs are because of very highly optimized libraries available for Nvidia cards (like CuDNN) and so on.
Re: TensorFlow 1.0 Released
#63Earlier quoted context omitted.
It's worth pointing out Tensorflow is basically Google's clone of Theano, including a lot of the same design decisions. They've improved some things but it's not like Google handed us the secret to fire here. It's just a good implementation of the same things a lot of people have been working on for years.
TensorFlow is not a clone of Theano. It's based on the earlier Google's platform DistBelief, mostly known outside of Google as the engine behind 2012 Youtube cat videos paper. Like DistBelief, TensorFlow was designed from the ground up to be scalable across multiple nodes. Theano, on the other hand, seems to be focused on the optimizations for the single machine, single GPU code. It only recently got the ability to r…
DistBelief was a CPU-only special purpose neural network system that would have been difficult to modify to support arbitrary neural architectures like theano. TensorFlow is not based on DistBelief in any meaningful way other than that they were written by mostly the same people.
Re: TensorFlow 1.0 Released
#64Is there a good birds-eye overview of what people are creating with TF or ML in general?
Re: TensorFlow 1.0 Released
#65Not necessarily about the article so may get downvoted, but is there a good book for TensorFlow/Machine Learning?
http://www.deeplearningbook.org/
Good book to learn the deep learning concepts. The official tensorflow tutorials are also good to learn the programming part which is not covered in the book.
Re: TensorFlow 1.0 Released
#66How do I get started with machine learning? I have a couple of applications in mind, mostly time series predictions. But the machine learning field seems to be vast and I don't know where to start.
Start with statistics. Seriously, just google time series modeling (this seems ok for a beginner https://www.analyticsvidhya.com/blog/2015/12/complete-tutori...). Learn ARMA/ARIMA/etc.
Don't worry that just because it isn't using deep nets that it isn't state of the art or won't get the job done well. That would be like thinking python's built-in sort function isn't sufficient because it doesn't use Spark.
Re: TensorFlow 1.0 Released
#67Kudos to the team. Anybody know if we can we train in languages other than Python yet (or do I have that wrong)?
there's an official "experimental" Java version on github, and many people are using and commiting to it
Re: TensorFlow 1.0 Released
#68Earlier quoted context omitted.
How do you embed in a mobile app?
Essentially there are two ways to do this. The “old” way is to export your TensorFlow neural network into a protobuf file, then load up the TensorFlow interpreter in your iOS/Android app, feed it the neural net, and run the inference directly on device. The GitHub repo [0] has a good set of examples of what that looks like in practice. The new, still experimental way is to compile your neural net into executable code…
Running neural nets on a standard mobile device will be game changing. I can't wait for mobile devices having custom chips to do AI related tasks.
I'm predicting in a decade we'll have offline speech and image recognition running on the phone.
Re: TensorFlow 1.0 Released
#69Earlier quoted context omitted.
Essentially there are two ways to do this. The “old” way is to export your TensorFlow neural network into a protobuf file, then load up the TensorFlow interpreter in your iOS/Android app, feed it the neural net, and run the inference directly on device. The GitHub repo [0] has a good set of examples of what that looks like in practice. The new, still experimental way is to compile your neural net into executable code…
Running neural nets on a standard mobile device will be game changing. I can't wait for mobile devices having custom chips to do AI related tasks. I'm predicting in a decade we'll have offline speech and image recognition running on the phone.
They don't want you to go there. Remind who started the latest big AI Projects (Google, Amazon, Microsoft, Facebook) and I don't think the will stop grabbing data.
I think they'll develop a hivemind, where mobile adds to the pool. In short Skynet ;)
Re: TensorFlow 1.0 Released
#70For a complete beginner. What kind of applications I can work on using TensorFlow?