installation via pip or conda
vaibhavnayel opened this issue · comments
Is there a way to use the package through pip/conda without docker? I'd also like to know what additional steps are needed to set this up in jupyter lab.
I'll second a request for this, there would be a ton of value in having this available as a pip installable package. @Roshrini is that on the roadmap for this plugin?
seconding this. So far the plugin seems very limited if we are forced to use it with an aws docker image.
thirding this, it would be great if we could get a pip installable package
for now you can try pip install like this without docker:
https://github.com/aws/deep-learning-containers/blob/master/pytorch/training/docker/1.9/py3/Dockerfile.cpu#L148-L149
ubuntu:~/code/amazon-s3-plugin-for-pytorch$ pip install -U https://aws-s3-plugin.s3.us-west-2.amazonaws.com/binaries/0.0.1/1c3e69e/awsio-0.0.1-cp38-cp38-manylinux1_x86_64.whl
ERROR: awsio-0.0.1-cp38-cp38-manylinux1_x86_64.whl is not a supported wheel on this platform.
WARNING: You are using pip version 21.3; however, version 21.3.1 is available.
You should consider upgrading via the '/home/ubuntu/Envs/doory/bin/python -m pip install --upgrade pip' command.
Can you update this to support 3.6
Nevermind, I found the address:
How do I do this if using a mac. When I try to pip install, it says platform not accepted.
Hi all, I've updated the installation guide for the plugin: https://github.com/aws/amazon-s3-plugin-for-pytorch#installation
However, we're deprecating support for the S3 plugin, and I suggest to try the new torchdata S3 IO datapipes. The S3 IO is upstreamed into torchdata package. We're supporting the new torchdata package, and continuously improving user experience and performance of the S3 IO datapipes.