clearlinux/dockerfiles
2
0
Fork 0
mirror of https://github.com/clearlinux/dockerfiles.git synced 2026-08-27 02:46:48 +00:00
Code Issues Packages Projects Releases Wiki Activity

updating official name to Intel Deep Learning As A Service Stack

Browse Source
This commit is contained in:
Beth Dean
2018-12-07 08:34:16 -08:00
committed by William Douglas
parent fd9fd44e82
commit 8f1b5284cd
1 changed files with 13 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files
stacks/dlaas/releasenote.md
+13 -13
View File
@@ -1,16 +1,16 @@
# Intel® Deep Learning Stack
# Intel® Deep Learning As A Service Stack
We are pleased to announce the initial release of the Intel® Deep Learning Stack.
We are pleased to announce the initial release of the Intel Deep Learning As A Service Stack.
The Intel® Deep Learning Stack is optimized for Intel Architecture (IA) from the hardware to the end-use applications and solutions. The stack has been tuned, tested and validated for Intel platforms to power a growing range of AI workloads. This stack makes it easier for developers and Cloud Service Providers to build deep learning models by reducing the complexity that comes with cutting-edge work.
The Intel Deep Learning As A Service Stack is a reference stack that is optimized for Intel Architecture (IA) from the hardware to the end-use applications and solutions. The stack has been tuned, tested and validated for Intel platforms to power a growing range of AI workloads. This reference stack makes it easier for developers and Cloud Service Providers to build deep learning models by reducing the complexity that comes with cutting-edge work.
# The Intel® Deep Learning Stack Release
# The Intel Deep Learning As A Service Stack Release
To offer more flexibility, we are releasing two versions of the Intel® Deep Learning Stack: a version that includes TensorFlow optimized for Intel® Architecture, the "Eigen" version and a second version that includes the TensorFlow* framework optimized using Intel® Math Kernel Library for Deep Neural Networks, the "Intel® MKL-DNN" version.
To offer more flexibility, we are releasing two versions of the Intel Deep Learning As A Service Stack: a version that includes TensorFlow* optimized for Intel® Architecture, the "Eigen" version, and a second version that includes the TensorFlow* framework optimized using Intel® Math Kernel Library for Deep Neural Networks, the "Intel® MKL-DNN" version.
> **Note:**
Clear Linux will be automatically updated to the latest release in the container. The minimum validated version of Clear Linux for this stack is 26240
@@ -18,7 +18,7 @@ To offer more flexibility, we are releasing two versions of the Intel® Deep Lea
> **Note:**
> For multi-node support, we include a registry with a set of jsonnet files to show integration with Kubeflow for deployment.
## Intel® Deep Learning Stack with Eigen
## The Intel Deep Learning As A Service Stack with Eigen
The release includes:
@@ -26,7 +26,7 @@ The release includes:
* TensorFlow 1.12 compiled with AVX2 and AVX512 optimizations
* Runtimes (python)
## Intel® Deep Learning Stack with Intel® MKL-DNN
## The Intel Deep Learning As A Service Stack with Intel® MKL-DNN
The release includes:
@@ -35,9 +35,9 @@ The release includes:
* TensorFlow 1.12 optimized using Intel® Math Kernel Library for Deep Neural Networks (Intel® MKL-DNN) primitives.
## How to get the Intel® Deep Learning Stack
## How to get the Intel Deep Learning As A Service Stack
The official Intel® Deep Learning Stack docker images are hosted at: https://hub.docker.com/u/clearlinux/.
The official Intel Deep Learning As A Service Stack docker images are hosted at: https://hub.docker.com/u/clearlinux/.
* Pull from the [Eigen version](https://hub.docker.com/r/clearlinux/stacks-dlaas-oss/)
* Pull from the [Intel MKL-DNN version](https://hub.docker.com/r/clearlinux/stacks-dlaas-mkl/)
@@ -48,14 +48,14 @@ The official Intel® Deep Learning Stack docker images are hosted at: https://hu
## Licensing
The Intel® Deep Learning Stack is guided by the same [Terms of Use](https://download.clearlinux.org/TermsOfUse.html) declared by the Clear Linux project. The Docker images are hosted on https://hub.docker.com and as with all Docker images, these likely also contain other software which may be under other licenses (such as Bash, etc. from the base distribution, along with any direct or indirect dependencies of the primary software being contained).
The Intel Deep Learning As A Service Stack is guided by the same [Terms of Use](https://download.clearlinux.org/TermsOfUse.html) declared by the Clear Linux project. The Docker images are hosted on https://hub.docker.com and as with all Docker images, these likely also contain other software which may be under other licenses (such as Bash, etc. from the base distribution, along with any direct or indirect dependencies of the primary software being contained).
# Working with the Intel® Deep Learning Stack
# Working with the Intel Deep Learning As A Service Stack
The Intel® Deep Learning Stack includes TensorFlow and Kubeflow support.
The Intel Deep Learning As A Service Stack includes TensorFlow and Kubeflow support.
These software components were selected because they are most popular/widely used by developers and CSPs. Clear Linux provides optimizations across the entire OS stack for the ultimate end user performance and is customizable to meet your unique needs. TensorFlow was selected as it is the leading deep learning and machine learning framework. Intel® Math Kernel Library for Deep Neural Networks (Intel® MKL-DNN) is an open source performance library for Deep Learning (DL) applications intended for acceleration of DL frameworks on Intel® architecture. Intel® MKL-DNN includes highly vectorized and threaded building blocks to implement convolutional neural networks (CNN) with C and C++ interfaces. Kubeflow is a project that provides a straightforward way to deploy simple, scalable and portable Machine Learning workflows on Kubernetes. This combination of an operating system, the deep learning framework and libraries, results in a performant deep learning software stack.
Please refer to the [Deep Learning tutorial](https://clearlinux.org/documentation/clear-linux/tutorials/dlaas.rst) for detailed instructions for running the TensorFlow and Kubeflow Benchmarks on the docker images.
@@ -133,7 +133,7 @@ Example: For Intel® Xeon® Gold 6140 CPU @ 2.30GHz based systems with 2 Sockets
For further notes on performance tuning use Tensorflow’s [official performance guide](https://www.tensorflow.org/guide/performance/overview)
# Contributing to the Intel® Deep Learning Stack
# Contributing to the Intel Deep Learning As A Service Stack
We encourage your contributions to this project, through the established Clear Linux community tools. Our team uses typical open source collaboration tools that are described on the Clear Linux [community page](https://clearlinux.org/community).