Rearrange files and projects to match #16 (#17)

.github/ISSUE_TEMPLATE.md

-## Expected Behavior
-
-
-## Actual Behavior
-
-
-## Steps to Reproduce the Problem
-
-1.
-1.
-1.
-
-## Specifications
-
-- Version:
-- Platform:
\ No newline at end of file

.github/PULL_REQUEST_TEMPLATE.md

-Fixes #<issue_number_goes_here>
-
-> It's a good idea to open an issue first for discussion.
-
-- [ ] Tests pass
-- [ ] Appropriate changes to README are included in PR
\ No newline at end of file

CODEOWNERS

+# SIG Leads are default for PRs unless a specific owner is listed below
+* @angerson @perfinion
+
+# These should match the Maintainer: line in your README.md
+# Please sort them in alphabetical order.
+/directory_template @angerson @perfinion
+/manylinux_2014_docker_images @sub-mod
+/ppc64le @wdirons
+/tekton @perfinion

CONTRIBUTING.md

 # How to Contribute
 
-We'd love to accept your patches and contributions to this project. There are
-just a few small guidelines you need to follow.
+To add your own project to our showcase, you can:
+
+* Add a link and a short description to [README.md](README.md) in the same
+  format as the other links, and then file a PR, or,
+* Copy the [Directory Template](directory_template) into a new folder, fill in
+  the template, and add it to the showcase. Make sure to update
+  [CODEOWNERS](CODEOWNERS) as well.
+
+This repo is a good place for small additions, but if the project is a big one,
+it may be better to maintain it in your own repository for better encapsulation
+of responsibility (and so you can make your own rules for further
+contributions).
+
+There are just a few small standard guidelines you need to follow as well:
 
 ## Contributor License Agreement
 

README.md

@@ -7,22 +7,49 @@
 [![Gitter chat](https://img.shields.io/badge/chat-on%20gitter-46bc99.svg)](https://gitter.im/tensorflow/sig-build)
 [![SIG Build Forum](https://img.shields.io/badge/discuss-on%20tensorflow.org-orange)](https://groups.google.com/a/tensorflow.org/g/build)
 
-**TensorFlow SIG Build** is a community group dedicated to improving all aspects
-of the TensorFlow build process.
+**TensorFlow SIG Build** is a community group dedicated to the TensorFlow build
+process. This repository is a showcase of resources, guides, tools, and builds
+contributed by the community, for the community.
 
-## Contributing
+## Group
 
-SIG Build is a community led open source project. As such, the project
+### Contributing
+
+SIG Build is a community-led open source project. As such, the project
 depends on public contributions, bug-fixes, and documentation. Please
 see [contribution guidelines](CONTRIBUTING.md) for a guide on how to
 contribute. This project adheres to [TensorFlow's code of conduct](CODE_OF_CONDUCT.md).
 By participating, you are expected to uphold this code.
 
-## Community
+### Community
 
 * [Public Mailing List](https://groups.google.com/a/tensorflow.org/forum/#!forum/build)
 * [SIG Monthly Meeting Notes](https://docs.google.com/document/d/10_3IQ5aF-88ADJNLF0WOpb09bZ15x-sBnRSnDHNCNr8/edit)
     * Join our mailing list and receive calendar invites to the meeting.
 
-## License
+### License
 [Apache License 2.0](LICENSE)
+
+## Project Showcase
+
+Want to add your own project to this list? It's easy: check out
+[CONTRIBUTING.md](CONTRIBUTING.md).
+
+### Docker
+
+* [**Manylinux 2014 Docker Images**](manylinux_2014_docker_images):
+  `manylinux2014` build environment for TensorFlow packages.
+* [**Distroless Dockerfiles**](https://github.com/uvarc/rivanna-docker):
+  Distroless ([info](https://github.com/GoogleContainerTools)) TensorFlow
+  images, which are smaller than TensorFlow's official images.
+
+### Platforms
+
+* [**ppc64le Builds**](ppc64le_builds): Dockerfiles and wheel build scripts for
+  building TF on ppc64le.
+
+### WIP / Other
+
+* [**Tekton CI**](tekton): perfinion's experimental directory for using Tekton 
+  CI with TensorFlow
+* [**Directory Template**](directory_template): Example short description.

directory_template/README.md

+# Directory Template 
+
+Example short description.
+
+Maintainer: @angerson (TensorFlow, SIG Build)
+
+* * *
+
+This is is a sample of a new directory for showcasing a local project. Anything
+goes, just please follow these rules to keep things straightforward:
+
+- Name the directory the same as the title `in_snake_case` (all lowercase, with
+  underscores, no hyphens).
+- Your directory should have a README.md with a header:
+    - Title identical to the one added on the root README.md
+    - Short description identical to the one added on the root README.md
+    - Maintainer: `<your email(s)> (<your organization(s)>)`. If you have no
+      organization or not acting on behalf of one, write "Independent".
+- You can add this directory to the root README.md list under any Showcase
+  section. If it doesn't fit under any section, just put it at the top under
+  "Showcase."

images/Dockerfile.linux.cpu

-FROM ubuntu:16.04
-
-LABEL maintainer="Jason Zaman <jason@perfinion.com>"
-
-# Copy and run each install script one at a time for better layer caching.
-
-COPY install/install_deb_packages.sh /install/
-RUN /install/install_deb_packages.sh
-
-COPY install/install_golang.sh /install/
-RUN /install/install_golang.sh
-
-COPY install/install_proto3.sh /install/
-RUN /install/install_proto3.sh
-
-COPY install/install_buildifier.sh /install/
-RUN /install/install_buildifier.sh
-
-COPY install/install_bazel.sh /install/
-RUN /install/install_bazel.sh
-
-COPY install/install_pip_packages.sh /install/
-RUN /install/install_pip_packages.sh
-
-# Set up the master bazelrc configuration file.
-COPY install/bazelrc /etc/bazel.bazelrc

images/README.md

-# Dockerfiles for TensorFlow builds
-
-## manylinux2014 builds
-Run the below commands to create manylinux2014 build environment.    
-```
-docker build -t build/manylinux2014 -f Dockerfile.manylinux2014.cpu .
-docker run -it build/manylinux2014:latest /bin/bash
-```
-
-Run the below commands to build TF.  
-```
-source scl_source enable devtoolset-7 rh-python36
-git clone --branch=r2.1 --depth=1 https://github.com/tensorflow/tensorflow.git
-cd tensorflow/
-./configure
-bazel build -c opt --cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" --verbose_failures //tensorflow/tools/pip_package:build_pip_package
-```
-
-## Distroless Images
-The [Distroless](https://github.com/GoogleContainerTools/distroless) base images developed by Google are meant to contain an application and its runtime dependencies only. Please visit their GitHub for more information.
-
-Distroless Dockerfiles for TensorFlow CPU (`Dockerfile.cpu`) and GPU (`Dockerfile.distroless`) are hosted in the [rivanna-docker repository](https://github.com/uvarc/rivanna-docker). Using the [MNIST benchmark](https://www.tensorflow.org/tutorials/quickstart/beginner), the containers were tested on UVA's HPC Rivanna cluster to have the same performance as the official ones.
-
-Image size comparison:
-| [Official](https://hub.docker.com/r/tensorflow/tensorflow/tags) | Size | [Distroless](https://hub.docker.com/r/uvarc/tensorflow/tags) | Size | Reduction |
-|---|---|---|---|---|
-| `2.3.0` | 582 MB | `2.3.0-cpu` | 227 MB | 61% |
-| `2.3.0-gpu` | 1.44 GB | `2.3.0-distroless` | 1.18 GB | 18% |
-| `nightly-gpu` | 2.27 GB | `2.4.0-distroless` | 1.47 GB | 35% |
-
-Pull command:
-```
-docker pull uvarc/tensorflow:<tag>
-```
-
-### Contact
-Research Computing at the University of Virginia (external) maintains Distroless Dockerfiles for TensorFlow in the [rivanna-docker repository](https://github.com/uvarc/rivanna-docker). Please reach out to hpc-support@virginia.edu for questions and feedback.

manylinux2014_docker_images/README.md

+# Manylinux2014 Docker Images
+
+`manylinux2014` build environment for TensorFlow packages.
+
+Maintainer: @sub-mod (Red Hat)
+
+* * *
+
+Run the below commands to create a manylinux2014 build environment.    
+```
+docker build -t build/manylinux2014 -f Dockerfile.manylinux2014.cpu .
+docker run -it build/manylinux2014:latest /bin/bash
+```
+
+Run the below commands to build TF.  
+```
+source scl_source enable devtoolset-7 rh-python36
+git clone --branch=r2.1 --depth=1 https://github.com/tensorflow/tensorflow.git
+cd tensorflow/
+./configure
+bazel build -c opt --cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" --verbose_failures //tensorflow/tools/pip_package:build_pip_package
+```
+

ppc64le_builds/README.md

+# ppc64le Builds
+
+Dockerfiles and wheel build scripts for building TF on ppc64le.
+
+Maintainer: @wdirons (IBM)
+
+* * *
+
+`cpu/Dockerfile.manylinux_2014` extends from the
+quay.io/pypa/manylinux2014_ppc64le docker image and installs java, bazel (from
+source), and the minimal required pip packages to build TensorFlow. It also adds
+the jenkins userid necessary to work with Oregon State University's Open Source
+Lab jenkins environment. `gpu/Dockerfile.manylinux_2014.cuda10_1` extends from
+the cpu image and adds CUDA 10.1 and cuDNN 7.6.
+
+The bash scripts are the equivalent of what is run in the TensorFlow builds done
+with Jenkins (everything except the remote cache server). They can also be run
+manually by cloning tensorflow and copying the script into the root of the
+workspace.