Update TensorFlow for SX-Aurora

 Posted on December 18, 2019  |  2 minutes  |  355 words  |  Kazuhisa Ishizaka

We have updated tensorflow-ve, Tensorflow with SX-Aurora TSUBASA Vector Engine Support. Now it supports tf 2.0 API! We release the prebuilt package. See README_ve.md for install instructions. Tensorflow-ve supports VectorEngine(VE) as a offload device such as GPU. That means our TF is VE-aware, for example you can see kernels run on VE and data transfer between CPU and VE on a standard profiler in tensorboard. And our TF includes some kernels that run on VE. [Read More]
tensorflow 

Deep Reinforcement Learning on SX-Aurora

 Posted on October 7, 2019  |  1 minutes  |  155 words  |  Kazuhisa Ishizaka

wellvill Co.,Ltd. has evaluated their deep reinforcement learning algorithm on SX-Aurora’s Vector Engine. The result is very impressive. SX-Aurora outperforms GPU(P100) system about two times. Their algorithm is based on “dueling double DQN” and solves 3D bin packing problems such as packing multiple products in boxes in a logistics center, loading packages into a truck, etc. Since they are using TensorFlow, it can run on VE with TensorFlow for VE without special modification. [Read More]
tensorflow  reinforcement learning 

Performance of TensorFlow on SX-Aurora

 Posted on July 8, 2019  |  2 minutes  |  294 words  |  Kazuhisa Ishizaka

We are investigating how SX-Aurora works on variouse ML applications. In this post, we would like to share the result of performance evaluation of three ML workloads using TensorFlow. The graphs show relative performance of training on CPU, GPU and VE in SX-Aurora. The left graph is simple CNN for image classification based on the example in Keras. We have used mnist dataset. As you know, GPU’s high peak computational performance works well for convolution layers, then V100 is the best. [Read More]
tensorflow  machine learning  logistic regression  ISC19 

Release TensorFlow for SX-Aurora

 Posted on June 3, 2019  |  1 minutes  |  179 words  |  Kazuhisa Ishizaka

We are pleased to announce the release of TensorFlow for SX-Aurora. This TF supports Vector Engin in SX-Aurora as a computing device. We have implemented some kernels for VE. Such kenrels are offloaded to VE for acceleration. We have also released: keras includes small modification for VE, vetfkernel includes implemetation of kernels for VE, and vednn is Vector Engine DNN Library. You can pip install prebuild packages to start to use TF on SX-Aurora. [Read More]
tensorflow