Fei Wang

Hi Anthony, Thanks for letting me know about TensorFlow Scala. It looks interesting. I just heard several days ago that someone are hoping for support to run machine learning models...

I am working on this but there is some problem with range selection. If I have apply(i: Rep[Int], j: Rep[Int]) to select a range, the (j-i+1) is actually a Rep[Int],...

Let's first be sure of the goal. Rep typed range/slicing allows slicing at runtime, but static range/slicing is only slicing at stage time. What is the goal? Is it slicing...

Runtime slicing is useful for RNN, whose input is 3D tensor of shape (timeStep, batch, dimension). At each cycle i, we process input(i) with the last hidden state, and produce...

Absolutely. I am glad that this repo has caught more attention from the community! I just added something to facilitate your request but there are still more complications. Let me...

Sorry about that. I totally agree with you on that issue. The ML community is overly centered on Python environment. Your situation poses an interesting challenge for me to see...

Not sure if you have found a solution to your problem, but I just learned that TensorFlow has Scala frontend (https://github.com/eaplatanios/tensorflow_scala). Could this help you?

is this branch still DoNotMerge?

Let's drop this for now. We should try to get small CNN case running on GPU ASAP

https://github.com/feiwang3311/Lantern/blob/master/src/main/scala/lantern/ad_lms_vector.scala#L522 @dan-zheng should this line be comparing this.shape(1) with that.shape(0)?