MXNet Extensions
Since I joined Amazon and started working on the Apache (incubating) MXNet open source deep learning framework I kept looking for ways to integrate hardware acceleration. The main problem is […]
Since I joined Amazon and started working on the Apache (incubating) MXNet open source deep learning framework I kept looking for ways to integrate hardware acceleration. The main problem is […]
Recently AWS announced the availability of ARM based EC2 instances. So I thought id try out compiling MXNet to run on one of these instances since MXNet has already been […]
Recently I had the need to calculate the diagonal of a matrix. Initially I thought, wow, here we go again: And initially this worked. But then I realized that it […]
I was in need of getting familiar with calling cuDNN routines, but the descriptor interface was a little confusing. So I put together this example calling the pooling operator for […]
This paper is a new direction looking at how we can take advantage of the C-based HLS tools that are currently available by translating Python functions to C so that […]
Python is great, and there are a lot of tools and libraries for all sorts of functionality available now. But sometimes you need to stitch tools together and use some […]
At the 2017 Reconfig conference I presented some work-in-progress on a set of Python tools that ive been developing to help with implementing Python Applications with FPGAs. A PDF version […]
I am an avid FPGA enthusiast and have worked with FPGAs for years, programming VHDL, Verilog, and C/C++ with High Level Synthesis (HLS) tools like Impulse-C and Vivado HLS. I […]
This paper is sort of an offshoot of our previous work on estimating the performance of computations using scheduling. To get a better understanding of the performance we started looking […]
Continuing with our previous work on heterogeneous systems, in this paper we describe the compilation framework to take an initial sequential Matlab program and parallelize it automatically across a CPU+GPU+FPGA system. […]