| Name | ck JSON |
| Version |
2.6.3
JSON |
| download |
| home_page | https://github.com/mlcommons/ck |
| Summary | Collective Knowledge - a lightweight knowledge manager to organize, cross-link, share and reuse artifacts and workflows based on FAIR principles |
| upload_time | 2022-11-30 10:15:41 |
| maintainer | |
| docs_url | None |
| author | Grigori Fursin |
| requires_python | |
| license | Apache 2.0 |
| keywords |
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
pyyaml
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
*Note that the 1st generation of the CK framework was discontinued in summer 2022
after the [2nd generation of this framework (CM)](https://github.com/mlcommons/ck)
was released by the [open taskforce on education and reproducibility](https://github.com/mlcommons/ck/blob/master/docs/mlperf-education-workgroup.md)
at [MLCommons](https://mlcommons.org).*
# Collective Knowledge framework (CK)
[](https://pepy.tech/project/ck)
[](https://badge.fury.io/py/ck)
[](https://pypi.org/project/ck)
[](https://travis-ci.com/ctuning/ck)
[](https://ci.appveyor.com/project/gfursin/ck)
[](https://coveralls.io/github/ctuning/ck)
[](https://ck.readthedocs.io/en/latest/?badge=latest)
[](https://colab.research.google.com/drive/1Fp6uxCqTazmCSSl8v-nY93VVmcOoLiXi?usp=sharing)
[](https://colab.research.google.com/drive/1a761nKgoHlJAy6gOXh-c9H4WkLV8nzRU?usp=sharing)
## News
* **2022 July 17:** We have pre-released CK2-based MLOps and DevOps automation scripts at https://github.com/mlcommons/ck/tree/master/cm-mlops/script
* **2022 May:** We started developing the 2nd generation of the CK framework (aka CM): https://github.com/mlcommons/ck/tree/master/cm
* **2022 April 3:** We presented the CK concept to bridge the growing gap between ML Systems research and production
at the HPCA'22 workshop on [benchmarking deep learning systems](https://sites.google.com/g.harvard.edu/mlperf-bench-hpca22/home).
* **2022 March:** We presented the [CK concept to enable collaborative and reproducible ML Systems R&D](https://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=73126)
at the SIAM'22 workshop on "Research Challenges and Opportunities within Software Productivity, Sustainability, and Reproducibility"
* **2022 March:** we've released the first prototype of [the Collective Mind toolkit (CK2)](https://github.com/mlcommons/ck/tree/master/ck2)
based on your feedback and our practical experience [reproducing 150+ ML and Systems papers and validating them in the real world](https://www.youtube.com/watch?v=7zpeIVwICa4).
## Motivation
While Machine Learning is becoming more and more important in everyday life, designing efficient ML Systems and deploying them in the real world
is becoming increasingly challenging, time consuming and costly.
Researchers and engineers must keep pace with rapidly evolving software stacks and a Cambrian explosion of hardware platforms from the cloud to the edge.
Such platforms have their own specific libraries, frameworks, APIs and specifications and often require repetitive, tedious and ad-hoc optimization
of the whole model/software/hardware stack to trade off accuracy, latency, throughout, power consumption, size and costs depending on user requirements and constraints.
### The CK framework
*The Collective Knowledge framework (CK)* is our attempt to develop a common plug&play infrastructure that can be used
by the community similar to Wikipedia to learn how to solve above challenges and make it easier to co-design,
benchmark, optimize and deploy Machine Learning Systems in the real world across continuously evolving software,
hardware and data sets (see our [ACM TechTalk](https://www.youtube.com/watch?v=7zpeIVwICa4) for more details):
* CK aims at providing a simple playground with minimal software dependencies to help researchers and practitioners share their knowledge
in the form of reusable automation recipes with a unified Python API, CLI and meta description:
- [Stable CK automation recipes](https://github.com/mlcommons/ck/tree/master/ck/repo/module)
- [MLPerf™ benchmark automation recipes](https://github.com/mlcommons/ck/tree/master/ck-mlops/repo/module)
* CK helps to organize software projects and Git repositories as a database of above automation recipes
and related artifacts based on [FAIR principles](https://www.nature.com/articles/sdata201618)
as described in our [journal article](https://arxiv.org/pdf/2011.01149.pdf) ([shorter pre-print]( https://arxiv.org/abs/2006.07161 )).
See examples of CK-compatible GitHub repositories:
- [MLPerf/MLOps automation](https://github.com/mlcommons/ck-mlops)
- [ACM REQUEST tournament for collaborative and reproducible ML/SW/HW co-design](https://github.com/ctuning/ck-request)
### Community developments
We collaborated with the community to reproduce [150+ ML and Systems papers](https://cTuning.org/ae)
and implement the following reusable automation recipes in the CK format:
* Portable meta package manager to automatically detect, install or rebuild various ML artifacts
(ML models, data sets, frameworks, libraries, etc) across different platform and operating systems including Linux, Windows, MacOS and Android:
- [ML artifact detection plugins](https://github.com/mlcommons/ck-mlops/tree/main/soft)
- [ML meta package installation plugins](https://github.com/mlcommons/ck-mlops/tree/main/package)
- OS descriptions: [Linux/MacOS/Android](https://github.com/mlcommons/ck-mlops/tree/main/os) ; [Windows](https://github.com/ctuning/ck-win/tree/main/os)
* Portable manager for Python virtual environments: [CK repo](https://github.com/mlcommons/ck-venv).
* Portable workflows to support collaborative, reproducible and cross-platform benchmarking:
- [ML Systems benchmarking](https://github.com/mlcommons/ck-mlops/tree/main/program)
- [Compiler benchmarking](https://github.com/ctuning/ctuning-programs/tree/master/program)
* Portable workflows to automate MLPerf™ benchmark:
- [End-to-end submission suite used by multiple organizations to automate the submission of MLPerf inference benchmark](https://github.com/mlcommons/ck/blob/master/docs/mlperf-automation/README.md)
- MLPerf inference v1.1 results: [MLCommons press-release](https://mlcommons.org/en/news/mlperf-inference-v11),
[Datacenter results](https://mlcommons.org/en/inference-datacenter-11),
[Edge results](https://mlcommons.org/en/inference-edge-11)
- [Reproducibility studies for MLPerf inference benchmark v1.1 automated by CK](https://github.com/mlcommons/ck/tree/master/docs/mlperf-automation/reproduce#reproducibility-reports-mlperf-inference-benchmark-v11)
- [Design space exploration of ML/SW/HW stacks and customizable visualization](https://cknowledge.io/result/crowd-benchmarking-mlperf-inference-classification-mobilenets-all)
Please contact [Grigori Fursin](https://www.linkedin.com/in/grigorifursin) if you are interested to join this community effort!
### Tutorials
* [CK automations for unified benchmarking]( https://colab.research.google.com/drive/1a761nKgoHlJAy6gOXh-c9H4WkLV8nzRU?usp=sharing )
* [CK-based MLPerf inference benchmark automation example]( https://colab.research.google.com/drive/1Fp6uxCqTazmCSSl8v-nY93VVmcOoLiXi?usp=sharing )
* [CK-based MLPerf inference vision benchmark v1.1 automation (TVM)]( https://colab.research.google.com/drive/1aywGlyD1ZRDtQTrQARVgL1882JcvmFK-?usp=sharing )
* [CK-based MLPerf inference vision benchmark v1.1 automation (ONNX)]( https://colab.research.google.com/drive/1ij1rWoqje5-Sn6UsdFj1OzYakudI2RIS?usp=sharing )
* [CK basics]( https://colab.research.google.com/drive/15lQgxuTSkEPqi4plaat1_v2gJcfIrATF?usp=sharing )
## Releases
### Development version
We are developing the 2nd generation of the CK framework (aka CM) based on your feedback:
* [CK2(CM) framework](https://github.com/mlcommons/ck/tree/master/cm)
* [CK2(CM) MLOps and DevOps automation scripts](https://github.com/mlcommons/ck/tree/master/cm-mlops)
### Stable versions
The latest version of the CK automation suite supported by MLCommons™:
* [CK framework v2.6.1 (Apache 2.0 license)](https://github.com/mlcommons/ck/releases/tag/V2.6.1)
* [CK automation suite for MLPerf™ and ML/SW/HW co-design](https://github.com/mlcommons/ck-mlops)
## Current projects
* [Automating MLPerf(tm) inference benchmark and packing ML models, data sets and frameworks as CK components with a unified API and meta description](https://github.com/mlcommons/ck/blob/master/docs/mlperf-automation/README.md)
* Developing customizable dashboards for MLPerf™ to help end-users select ML/SW/HW stacks on a Pareto frontier: [aggregated MLPerf™ results]( https://cknowledge.io/?q="mlperf-inference-all" )
* Providing a common format to share artifacts at ML, systems and other conferences: [video](https://youtu.be/DIkZxraTmGM), [Artifact Evaluation](https://cTuning.org/ae)
* Redesigning CK together with the community based on user feedback: [incubator](https://github.com/mlcommons/ck/tree/master/incubator)
* [Other real-world use cases](https://cKnowledge.org/partners.html) from MLPerf™, Qualcomm, Arm, General Motors, IBM, the Raspberry Pi foundation, ACM and other great partners;
## Documentation
* [Online CK documentation]( https://ck.readthedocs.io )
* [Why CK?]( https://ck.readthedocs.io/en/latest/src/introduction.html )
* [CK Basics](https://michel-steuwer.github.io/About-CK)
* [Try CK]( https://ck.readthedocs.io/en/latest/src/first-steps.html )
* [Publications](https://github.com/mlcommons/ck/wiki/Publications)
## Installation
Follow [this guide](https://ck.readthedocs.io/en/latest/src/installation.html)
to install CK framework on your platform.
CK supports the following platforms:
| | As a host platform | As a target platform |
|---------------|:------------------:|:--------------------:|
| Generic Linux | ✓ | ✓ |
| Linux (Arm) | ✓ | ✓ |
| Raspberry Pi | ✓ | ✓ |
| MacOS | ✓ | ± |
| Windows | ✓ | ✓ |
| Android | ± | ✓ |
| iOS | TBD | TBD |
| Bare-metal (edge devices) | - | ± |
## Examples
### Portable CK workflow (native environment without Docker)
Here we show how to pull a GitHub repo in the CK format
and use a unified CK interface to compile and run
any program (image corner detection in our case)
with any compatible data set on any compatible platform:
```bash
python3 -m pip install ck
ck pull repo:mlcommons@ck-mlops
ck ls program:*susan*
ck search dataset --tags=jpeg
ck pull repo:ctuning-datasets-min
ck search dataset --tags=jpeg
ck detect soft:compiler.gcc
ck detect soft:compiler.llvm
ck show env --tags=compiler
ck compile program:image-corner-detection --speed
ck run program:image-corner-detection --repeat=1 --env.MY_ENV=123 --env.TEST=xyz
```
You can check output of this program in the following directory:
```bash
cd `ck find program:image-corner-detection`/tmp
ls
processed-image.pgm
```
You can now view this image with detected corners.
Check [CK docs](https://ck.readthedocs.io/en/latest/src/introduction.html) for further details.
### MLPerf™ benchmark workflows
* [Current coverage](https://github.com/mlcommons/ck/tree/master/docs/mlperf-automation#readme)
* [MLPerf inference v1.1 workflows](https://github.com/mlcommons/ck/blob/master/docs/mlperf-automation/reproduce/README.md#reproducibility-reports-mlperf-inference-benchmark-v11)
### Portable CK workflows inside containers
We have prepared adaptive CK containers to demonstrate MLOps capabilities:
* https://github.com/mlcommons/ck-mlops/tree/main/docker
You can run them as follows:
```bash
ck pull repo:mlcommons@ck-mlops
ck build docker:ck-template-mlperf --tag=ubuntu-20.04
ck run docker:ck-template-mlperf --tag=ubuntu-20.04
```
### Portable workflow example with virtual CK environments
You can create multiple [virtual CK environments](https://github.com/mlcommons/ck-venv) with templates
to automatically install different CK packages and workflows, for example for MLPerf™ inference:
```
ck pull repo:mlcommons@ck-venv
ck create venv:test --template=mlperf-inference-main
ck ls venv
ck activate venv:test
ck pull repo:mlcommons@ck-mlops
ck install package --ask --tags=dataset,coco,val,2017,full
ck show env
```
### Integration with web services and CI platforms
All CK modules, automation actions and workflows are accessible as a micro-service with a unified JSON I/O API
to make it easier to integrate them with web services and CI platforms as described
[here](https://github.com/mlcommons/ck/blob/master/docs/mlperf-automation/tools/continuous-integration.md).
### Other use cases
* [List of various use cases]( https://ck.readthedocs.io/en/latest/src/introduction.html#ck-showroom )
## CK portal
We have developed the [cKnowledge.io portal](https://cKnowledge.io) to help the community
organize and find all the CK workflows and components similar to PyPI:
* [Search CK components](https://cKnowledge.io)
* [Browse CK components](https://cKnowledge.io/browse)
* [Find reproduced results from papers]( https://cKnowledge.io/reproduced-results )
* [Test CK workflows to benchmark and optimize ML Systems]( https://cKnowledge.io/demo )
## Containers to test CK automation recipes and workflows
The community provides Docker containers to test CK and components using different ML/SW/HW stacks (DSE).
* A set of Docker containers to test the basic CK functionality
using some MLPerf inference benchmark workflows:
https://github.com/mlcommons/ck-mlops/tree/main/docker/test-ck
## Contributions
Users can extend the CK functionality via [CK modules](https://github.com/mlcommons/ck/tree/master/ck/repo/module)
or external [GitHub reposities](https://cKnowledge.io/repos) in the CK format
as described [here](https://ck.readthedocs.io/en/latest/src/typical-usage.html).
Please check [this documentation](https://ck.readthedocs.io/en/latest/src/how-to-contribute.html)
if you want to extend the CK core functionality and [modules](https://github.com/mlcommons/ck/tree/master/ck/repo/module).
Note, that we plan to [redesign the CK framework](https://github.com/mlcommons/ck/projects/1)
to be more pythonic (we wrote the first prototype without OO to be able
to port it to bare-metal devices in C but eventually we decided to drop this idea).
Please contact [Grigori Fursin](mailto:grigori@octoml.ai) to join this community effort.
## Contacts
* [Grigori Fursin](https://fursin.net) - author
* [Arjun Suresh](https://www.linkedin.com/in/arjunsuresh) - maintainer
## Acknowledgments
We would like to thank all [contributors](https://github.com/mlcommons/ck/blob/master/CONTRIBUTING.md)
and [collaborators](https://cKnowledge.org/partners.html) for their support, fruitful discussions,
and useful feedback! See more acknowledgments in the [CK journal article](https://arxiv.org/abs/2011.01149)
and [ACM TechTalk](https://www.youtube.com/watch?v=7zpeIVwICa4).
Raw data
{
"_id": null,
"home_page": "https://github.com/mlcommons/ck",
"name": "ck",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "",
"author": "Grigori Fursin",
"author_email": "Grigori.Fursin@cTuning.org",
"download_url": "https://files.pythonhosted.org/packages/f4/29/b4c314545e53f113816b2965ce6b457272dff06d7ba4a346ad1196cb60cb/ck-2.6.3.tar.gz",
"platform": null,
"description": "*Note that the 1st generation of the CK framework was discontinued in summer 2022 \n after the [2nd generation of this framework (CM)](https://github.com/mlcommons/ck) \n was released by the [open taskforce on education and reproducibility](https://github.com/mlcommons/ck/blob/master/docs/mlperf-education-workgroup.md)\n at [MLCommons](https://mlcommons.org).*\n\n\n# Collective Knowledge framework (CK)\n\n[](https://pepy.tech/project/ck)\n[](https://badge.fury.io/py/ck)\n[](https://pypi.org/project/ck)\n\n[](https://travis-ci.com/ctuning/ck)\n[](https://ci.appveyor.com/project/gfursin/ck)\n[](https://coveralls.io/github/ctuning/ck)\n\n[](https://ck.readthedocs.io/en/latest/?badge=latest)\n[](https://colab.research.google.com/drive/1Fp6uxCqTazmCSSl8v-nY93VVmcOoLiXi?usp=sharing)\n[](https://colab.research.google.com/drive/1a761nKgoHlJAy6gOXh-c9H4WkLV8nzRU?usp=sharing)\n\n\n## News\n\n\n\n* **2022 July 17:** We have pre-released CK2-based MLOps and DevOps automation scripts at https://github.com/mlcommons/ck/tree/master/cm-mlops/script\n\n* **2022 May:** We started developing the 2nd generation of the CK framework (aka CM): https://github.com/mlcommons/ck/tree/master/cm\n\n* **2022 April 3:** We presented the CK concept to bridge the growing gap between ML Systems research and production \n at the HPCA'22 workshop on [benchmarking deep learning systems](https://sites.google.com/g.harvard.edu/mlperf-bench-hpca22/home).\n\n* **2022 March:** We presented the [CK concept to enable collaborative and reproducible ML Systems R&D](https://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=73126) \n at the SIAM'22 workshop on \"Research Challenges and Opportunities within Software Productivity, Sustainability, and Reproducibility\"\n\n* **2022 March:** we've released the first prototype of [the Collective Mind toolkit (CK2)](https://github.com/mlcommons/ck/tree/master/ck2)\n based on your feedback and our practical experience [reproducing 150+ ML and Systems papers and validating them in the real world](https://www.youtube.com/watch?v=7zpeIVwICa4).\n\n## Motivation\n\nWhile Machine Learning is becoming more and more important in everyday life, designing efficient ML Systems and deploying them in the real world \nis becoming increasingly challenging, time consuming and costly.\nResearchers and engineers must keep pace with rapidly evolving software stacks and a Cambrian explosion of hardware platforms from the cloud to the edge. \nSuch platforms have their own specific libraries, frameworks, APIs and specifications and often require repetitive, tedious and ad-hoc optimization \nof the whole model/software/hardware stack to trade off accuracy, latency, throughout, power consumption, size and costs depending on user requirements and constraints.\n\n### The CK framework\n\n*The Collective Knowledge framework (CK)* is our attempt to develop a common plug&play infrastructure that can be used\nby the community similar to Wikipedia to learn how to solve above challenges and make it easier to co-design, \nbenchmark, optimize and deploy Machine Learning Systems in the real world across continuously evolving software, \nhardware and data sets (see our [ACM TechTalk](https://www.youtube.com/watch?v=7zpeIVwICa4) for more details):\n\n* CK aims at providing a simple playground with minimal software dependencies to help researchers and practitioners share their knowledge\n in the form of reusable automation recipes with a unified Python API, CLI and meta description: \n - [Stable CK automation recipes](https://github.com/mlcommons/ck/tree/master/ck/repo/module)\n - [MLPerf™ benchmark automation recipes](https://github.com/mlcommons/ck/tree/master/ck-mlops/repo/module)\n\n* CK helps to organize software projects and Git repositories as a database of above automation recipes \n and related artifacts based on [FAIR principles](https://www.nature.com/articles/sdata201618)\n as described in our [journal article](https://arxiv.org/pdf/2011.01149.pdf) ([shorter pre-print]( https://arxiv.org/abs/2006.07161 )).\n See examples of CK-compatible GitHub repositories: \n - [MLPerf/MLOps automation](https://github.com/mlcommons/ck-mlops)\n - [ACM REQUEST tournament for collaborative and reproducible ML/SW/HW co-design](https://github.com/ctuning/ck-request)\n\n### Community developments\n\nWe collaborated with the community to reproduce [150+ ML and Systems papers](https://cTuning.org/ae)\nand implement the following reusable automation recipes in the CK format: \n\n* Portable meta package manager to automatically detect, install or rebuild various ML artifacts \n (ML models, data sets, frameworks, libraries, etc) across different platform and operating systems including Linux, Windows, MacOS and Android:\n - [ML artifact detection plugins](https://github.com/mlcommons/ck-mlops/tree/main/soft)\n - [ML meta package installation plugins](https://github.com/mlcommons/ck-mlops/tree/main/package)\n - OS descriptions: [Linux/MacOS/Android](https://github.com/mlcommons/ck-mlops/tree/main/os) ; [Windows](https://github.com/ctuning/ck-win/tree/main/os) \n\n* Portable manager for Python virtual environments: [CK repo](https://github.com/mlcommons/ck-venv).\n\n* Portable workflows to support collaborative, reproducible and cross-platform benchmarking:\n - [ML Systems benchmarking](https://github.com/mlcommons/ck-mlops/tree/main/program)\n - [Compiler benchmarking](https://github.com/ctuning/ctuning-programs/tree/master/program)\n\n* Portable workflows to automate MLPerf™ benchmark:\n - [End-to-end submission suite used by multiple organizations to automate the submission of MLPerf inference benchmark](https://github.com/mlcommons/ck/blob/master/docs/mlperf-automation/README.md)\n - MLPerf inference v1.1 results: [MLCommons press-release](https://mlcommons.org/en/news/mlperf-inference-v11), \n [Datacenter results](https://mlcommons.org/en/inference-datacenter-11), \n [Edge results](https://mlcommons.org/en/inference-edge-11)\n - [Reproducibility studies for MLPerf inference benchmark v1.1 automated by CK](https://github.com/mlcommons/ck/tree/master/docs/mlperf-automation/reproduce#reproducibility-reports-mlperf-inference-benchmark-v11)\n - [Design space exploration of ML/SW/HW stacks and customizable visualization](https://cknowledge.io/result/crowd-benchmarking-mlperf-inference-classification-mobilenets-all)\n\n\nPlease contact [Grigori Fursin](https://www.linkedin.com/in/grigorifursin) if you are interested to join this community effort!\n\n### Tutorials\n\n* [CK automations for unified benchmarking]( https://colab.research.google.com/drive/1a761nKgoHlJAy6gOXh-c9H4WkLV8nzRU?usp=sharing )\n* [CK-based MLPerf inference benchmark automation example]( https://colab.research.google.com/drive/1Fp6uxCqTazmCSSl8v-nY93VVmcOoLiXi?usp=sharing )\n * [CK-based MLPerf inference vision benchmark v1.1 automation (TVM)]( https://colab.research.google.com/drive/1aywGlyD1ZRDtQTrQARVgL1882JcvmFK-?usp=sharing )\n * [CK-based MLPerf inference vision benchmark v1.1 automation (ONNX)]( https://colab.research.google.com/drive/1ij1rWoqje5-Sn6UsdFj1OzYakudI2RIS?usp=sharing )\n* [CK basics]( https://colab.research.google.com/drive/15lQgxuTSkEPqi4plaat1_v2gJcfIrATF?usp=sharing )\n\n## Releases\n\n### Development version\n\nWe are developing the 2nd generation of the CK framework (aka CM) based on your feedback:\n* [CK2(CM) framework](https://github.com/mlcommons/ck/tree/master/cm)\n* [CK2(CM) MLOps and DevOps automation scripts](https://github.com/mlcommons/ck/tree/master/cm-mlops)\n\n### Stable versions\n\nThe latest version of the CK automation suite supported by MLCommons™:\n* [CK framework v2.6.1 (Apache 2.0 license)](https://github.com/mlcommons/ck/releases/tag/V2.6.1)\n* [CK automation suite for MLPerf™ and ML/SW/HW co-design](https://github.com/mlcommons/ck-mlops)\n\n\n## Current projects\n* [Automating MLPerf(tm) inference benchmark and packing ML models, data sets and frameworks as CK components with a unified API and meta description](https://github.com/mlcommons/ck/blob/master/docs/mlperf-automation/README.md)\n* Developing customizable dashboards for MLPerf™ to help end-users select ML/SW/HW stacks on a Pareto frontier: [aggregated MLPerf™ results]( https://cknowledge.io/?q=\"mlperf-inference-all\" )\n* Providing a common format to share artifacts at ML, systems and other conferences: [video](https://youtu.be/DIkZxraTmGM), [Artifact Evaluation](https://cTuning.org/ae)\n* Redesigning CK together with the community based on user feedback: [incubator](https://github.com/mlcommons/ck/tree/master/incubator)\n* [Other real-world use cases](https://cKnowledge.org/partners.html) from MLPerf™, Qualcomm, Arm, General Motors, IBM, the Raspberry Pi foundation, ACM and other great partners;\n\n## Documentation\n\n* [Online CK documentation]( https://ck.readthedocs.io ) \n * [Why CK?]( https://ck.readthedocs.io/en/latest/src/introduction.html ) \n * [CK Basics](https://michel-steuwer.github.io/About-CK)\n * [Try CK]( https://ck.readthedocs.io/en/latest/src/first-steps.html )\n* [Publications](https://github.com/mlcommons/ck/wiki/Publications)\n\n## Installation\n\nFollow [this guide](https://ck.readthedocs.io/en/latest/src/installation.html) \nto install CK framework on your platform.\n\nCK supports the following platforms:\n\n| | As a host platform | As a target platform |\n|---------------|:------------------:|:--------------------:|\n| Generic Linux | \u2713 | \u2713 |\n| Linux (Arm) | \u2713 | \u2713 |\n| Raspberry Pi | \u2713 | \u2713 |\n| MacOS | \u2713 | \u00b1 |\n| Windows | \u2713 | \u2713 |\n| Android | \u00b1 | \u2713 |\n| iOS | TBD | TBD |\n| Bare-metal (edge devices) | - | \u00b1 |\n\n## Examples\n\n### Portable CK workflow (native environment without Docker)\n\nHere we show how to pull a GitHub repo in the CK format \nand use a unified CK interface to compile and run \nany program (image corner detection in our case)\nwith any compatible data set on any compatible platform:\n\n```bash\npython3 -m pip install ck\n\nck pull repo:mlcommons@ck-mlops\n\nck ls program:*susan*\n\nck search dataset --tags=jpeg\n\nck pull repo:ctuning-datasets-min\n\nck search dataset --tags=jpeg\n\nck detect soft:compiler.gcc\nck detect soft:compiler.llvm\n\nck show env --tags=compiler\n\nck compile program:image-corner-detection --speed\n\nck run program:image-corner-detection --repeat=1 --env.MY_ENV=123 --env.TEST=xyz\n```\n\nYou can check output of this program in the following directory:\n```bash\ncd `ck find program:image-corner-detection`/tmp\nls\n\nprocessed-image.pgm\n```\n\nYou can now view this image with detected corners.\n\n\nCheck [CK docs](https://ck.readthedocs.io/en/latest/src/introduction.html) for further details.\n\n### MLPerf™ benchmark workflows\n\n* [Current coverage](https://github.com/mlcommons/ck/tree/master/docs/mlperf-automation#readme)\n* [MLPerf inference v1.1 workflows](https://github.com/mlcommons/ck/blob/master/docs/mlperf-automation/reproduce/README.md#reproducibility-reports-mlperf-inference-benchmark-v11)\n\n### Portable CK workflows inside containers\n\nWe have prepared adaptive CK containers to demonstrate MLOps capabilities:\n* https://github.com/mlcommons/ck-mlops/tree/main/docker\n\nYou can run them as follows:\n\n```bash\nck pull repo:mlcommons@ck-mlops\nck build docker:ck-template-mlperf --tag=ubuntu-20.04\nck run docker:ck-template-mlperf --tag=ubuntu-20.04\n```\n\n### Portable workflow example with virtual CK environments\n\nYou can create multiple [virtual CK environments](https://github.com/mlcommons/ck-venv) with templates\nto automatically install different CK packages and workflows, for example for MLPerf™ inference:\n\n```\nck pull repo:mlcommons@ck-venv\nck create venv:test --template=mlperf-inference-main\nck ls venv\nck activate venv:test\n\nck pull repo:mlcommons@ck-mlops\nck install package --ask --tags=dataset,coco,val,2017,full\nck show env\n\n```\n\n### Integration with web services and CI platforms\n\nAll CK modules, automation actions and workflows are accessible as a micro-service with a unified JSON I/O API\nto make it easier to integrate them with web services and CI platforms as described \n[here](https://github.com/mlcommons/ck/blob/master/docs/mlperf-automation/tools/continuous-integration.md).\n\n\n\n\n### Other use cases\n\n* [List of various use cases]( https://ck.readthedocs.io/en/latest/src/introduction.html#ck-showroom )\n\n\n\n\n\n\n\n## CK portal \n\nWe have developed the [cKnowledge.io portal](https://cKnowledge.io) to help the community\norganize and find all the CK workflows and components similar to PyPI:\n\n* [Search CK components](https://cKnowledge.io)\n* [Browse CK components](https://cKnowledge.io/browse)\n* [Find reproduced results from papers]( https://cKnowledge.io/reproduced-results )\n* [Test CK workflows to benchmark and optimize ML Systems]( https://cKnowledge.io/demo )\n\n\n\n## Containers to test CK automation recipes and workflows\n\nThe community provides Docker containers to test CK and components using different ML/SW/HW stacks (DSE).\n\n* A set of Docker containers to test the basic CK functionality\n using some MLPerf inference benchmark workflows: \n https://github.com/mlcommons/ck-mlops/tree/main/docker/test-ck\n\n\n## Contributions\n\nUsers can extend the CK functionality via [CK modules](https://github.com/mlcommons/ck/tree/master/ck/repo/module) \nor external [GitHub reposities](https://cKnowledge.io/repos) in the CK format\nas described [here](https://ck.readthedocs.io/en/latest/src/typical-usage.html).\n\nPlease check [this documentation](https://ck.readthedocs.io/en/latest/src/how-to-contribute.html)\nif you want to extend the CK core functionality and [modules](https://github.com/mlcommons/ck/tree/master/ck/repo/module). \n\nNote, that we plan to [redesign the CK framework](https://github.com/mlcommons/ck/projects/1) \nto be more pythonic (we wrote the first prototype without OO to be able \nto port it to bare-metal devices in C but eventually we decided to drop this idea).\n\nPlease contact [Grigori Fursin](mailto:grigori@octoml.ai) to join this community effort.\n\n\n\n## Contacts\n\n* [Grigori Fursin](https://fursin.net) - author\n* [Arjun Suresh](https://www.linkedin.com/in/arjunsuresh) - maintainer\n\n## Acknowledgments\n\nWe would like to thank all [contributors](https://github.com/mlcommons/ck/blob/master/CONTRIBUTING.md) \nand [collaborators](https://cKnowledge.org/partners.html) for their support, fruitful discussions, \nand useful feedback! See more acknowledgments in the [CK journal article](https://arxiv.org/abs/2011.01149)\nand [ACM TechTalk](https://www.youtube.com/watch?v=7zpeIVwICa4).",
"bugtrack_url": null,
"license": "Apache 2.0",
"summary": "Collective Knowledge - a lightweight knowledge manager to organize, cross-link, share and reuse artifacts and workflows based on FAIR principles",
"version": "2.6.3",
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"md5": "fcef9456e0a6f82e335f6428f71aa96d",
"sha256": "227737f4b99af2f18d54560caa1099106fe1d7290ded2c969a0fe6c746ef9ce8"
},
"downloads": -1,
"filename": "ck-2.6.3.tar.gz",
"has_sig": false,
"md5_digest": "fcef9456e0a6f82e335f6428f71aa96d",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 1024267,
"upload_time": "2022-11-30T10:15:41",
"upload_time_iso_8601": "2022-11-30T10:15:41.118257Z",
"url": "https://files.pythonhosted.org/packages/f4/29/b4c314545e53f113816b2965ce6b457272dff06d7ba4a346ad1196cb60cb/ck-2.6.3.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2022-11-30 10:15:41",
"github": true,
"gitlab": false,
"bitbucket": false,
"github_user": "mlcommons",
"github_project": "ck",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"requirements": [
{
"name": "pyyaml",
"specs": []
}
],
"lcname": "ck"
}
