Don't hesitate to get in touch if you encounter any issues or would like to discuss this community project!
Please report if this CK component works: 1▲
or fails: 0▲
Sign up to be notified when artifacts are shared or updated!
Description
This meta package is our attempt to provide a unified Python API, CLI
and JSON meta description
for different package managers and building tools
to automatically download and install different components (models, data sets, libraries, frameworks, tools)
necessary to run portable program pipelines
across evolving platforms.
Our on-going project is to make the onboarding process as simple as possible via this platform.
Please check this CK white paper
and don't hesitate to contact us if you have suggestions or feedback!
Unfortunately, ImageNet 2012 validation set is no longer publicly available and this CK meta-package can not automatically download it!
See the related discussions:
* https://CodeReef.ai/portal/c/package/imagenet-2012-val
* https://github.com/pytorch/vision/issues/1453
* https://github.com/ctuning/ck-env/issues/101
However, if you already have it installed on your machine, you can detect and register it to work with CK workflows using this command:
$ ck detect soft:dataset.imagenet.train --search_dir={directory where the dataset is installed}
Maybe we can add a key to "meta.json" something like "force_license_accept":"yes" ?
In such case, CK will print the LICENSE.txt and ask a user to accept the license before installing this package?
Any thoughts?
Subject: about Imagenet licence
Date: Wed, 11 Dec 2019 21:46:00 -0800 (PST)
From: 'Thai Nguyen' via MLPerf Reply-To: Thai Nguyen To: MLPerf
Hi all,
I am contributing to the mlperf mobile app (https://github.com/mlperf/mobile_app).
An existing problem is users have to download and push the Imagenet dataset to the device themselves.
This is due to the Imagenet license (http://image-net.org/download-faq) which does not allow to publicly redistribute the dataset.
However, observing the CK's way of downloading the dataset. I saw that the users just need to run one command:
$ ck install package --tags=image-classification,dataset,imagenet,val,original,full
Look like the users don't need to involve in any license-related thing before downloading the dataset.
I wonder what is the legal (license) basis behind this?
Can I do such similar things to help user get the dataset easier in the app? May be:
- Show a pop-up that display the license
- Let them agree and then download in in the same way as CK
See the related discussions:
* https://CodeReef.ai/portal/c/package/imagenet-2012-val
* https://github.com/pytorch/vision/issues/1453
* https://github.com/ctuning/ck-env/issues/101
However, if you already have it installed on your machine, you can detect and register it to work with CK workflows using this command:
$ ck detect soft:dataset.imagenet.train --search_dir={directory where the dataset is installed}
In such case, CK will print the LICENSE.txt and ask a user to accept the license before installing this package?
Any thoughts?
Date: Wed, 11 Dec 2019 21:46:00 -0800 (PST)
From: 'Thai Nguyen' via MLPerf
Reply-To: Thai Nguyen
To: MLPerf
Hi all,
I am contributing to the mlperf mobile app (https://github.com/mlperf/mobile_app).
An existing problem is users have to download and push the Imagenet dataset to the device themselves.
This is due to the Imagenet license (http://image-net.org/download-faq) which does not allow to publicly redistribute the dataset.
However, observing the CK's way of downloading the dataset. I saw that the users just need to run one command:
$ ck install package --tags=image-classification,dataset,imagenet,val,original,full
Look like the users don't need to involve in any license-related thing before downloading the dataset.
I wonder what is the legal (license) basis behind this?
Can I do such similar things to help user get the dataset easier in the app? May be:
- Show a pop-up that display the license
- Let them agree and then download in in the same way as CK
Thanks.
Thai