FazBrowse GitHub Viewer | Trending |
URL:
| Home
Tools: [Download Repo ZIP]   [Original HTTPS Page]

Unpin MxNet by lopuhin · Pull Request #169 · Kaggle/docker-python · GitHub

Repository navigation

Unpin MxNet - #169

Merged
nerdcha merged 2 commits into
Kaggle:masterfrom
lopuhin:unpin-mxnet
Jan 28, 2018
Merged

nerdcha merged 2 commits into
Kaggle:masterfrom
lopuhin:unpin-mxnet

Conversation

lopuhin commented Jan 24, 2018

Copy link
Copy Markdown
Contributor

This unpins mxnet version. It was build from source before, and then moved to pip install in 6c4cac1.

I verified that package is installed correctly by building the docker image locally. Here is the relevant part of the build log where the package is installed:

Collecting mxnet
  Downloading mxnet-1.0.0.post4-py2.py3-none-manylinux1_x86_64.whl (27.4MB)
Requirement already satisfied: requests==2.18.4 in /opt/conda/lib/python3.6/site-packages (from mxnet)
Collecting graphviz==0.8.1 (from mxnet)
  Downloading graphviz-0.8.1-py2.py3-none-any.whl
Requirement already satisfied: numpy<=1.13.3 in /opt/conda/lib/python3.6/site-packages (from mxnet)
Requirement already satisfied: chardet<3.1.0,>=3.0.2 in /opt/conda/lib/python3.6/site-packages (from requests==2.18.4->mxnet)
Requirement already satisfied: idna<2.7,>=2.5 in /opt/conda/lib/python3.6/site-packages (from requests==2.18.4->mxnet)
Requirement already satisfied: urllib3<1.23,>=1.21.1 in /opt/conda/lib/python3.6/site-packages (from requests==2.18.4->mxnet)
Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.6/site-packages (from requests==2.18.4->mxnet)
Installing collected packages: graphviz, mxnet
Successfully installed graphviz-0.8.1 mxnet-1.0.0.post4

and other major libraries seems to be fine as well.

sebbov left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Choose a reason Spam Abuse Off Topic Outdated Duplicate Resolved Low Quality

Thanks Konstantin!

Anything you could also contribute to test mxnet invariants (even if just importing the right module(s)) in https://github.com/Kaggle/docker-python/blob/master/test_build.py ?

Otherwise this lgtm, but I'll let @nerdcha approve since he may be aware of a reason for the existing pinning.

sebbov requested a review from nerdcha January 25, 2018 18:28

lopuhin commented Jan 25, 2018

Copy link
Copy Markdown
Contributor Author

Thanks @sebbov , I'll rebase on master and double-check that mxnet is still installed correctly, and will add a check to test_build.py.

This unpins mxnet version. It was build from source before, and then moved to pip install in 6c4cac1.

lopuhin commented Jan 25, 2018

Copy link
Copy Markdown
Contributor Author

@nerdcha double-checked that everything is ok, rebased on master and added check that mxnet and mxnet.gluon can be imported to test_build.py.

nerdcha commented Jan 28, 2018

Copy link
Copy Markdown
Contributor

I'd pinned the version because I was following the practice from MXNet's docs, no other reason. This is much better. Thanks very much!

nerdcha merged commit fbb04f9 into Kaggle:master Jan 28, 2018
lopuhin deleted the unpin-mxnet branch January 29, 2018 06:58
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants


Back | FazBrowse Home | New Git URL