Open nick-youngblut opened 2 years ago
Thanks for your input. I appreciate feedback from users. I will work on a more user-friendly interface and release it asap.
On Sat, Mar 5, 2022 at 8:02 AM Nick Youngblut @.***> wrote:
The Deeplasmid Docker container for GPU section in the README states to use the following command for running deeplasmid for plasmid identification on GPU:
sudo /usr/bin/docker run -it --rm $(ls /dev/nvidia | xargs -I{} echo '--device={}') $(ls /usr/lib/-linux-gnu/{libcuda,libnvidia}* | \
xargs -I{} echo '-v {}:{}:ro') -v
pwd
/testing/649989979/649989979.fna:/srv/jgi-ml/classifier/dl/in.fasta \-v
pwd
/testing/649989979/649989979.fna.OUT:/srv/jgi-ml/classifier/dl/outdir \billandreo/deeplasmid-gpu feature_DL_plasmid_predict.sh in.fasta outdir
It would be great to have a more user-friendly interface, such as a bash/python script with arg parsing
— Reply to this email directly, view it on GitHub https://github.com/wandreopoulos/deeplasmid/issues/2, or unsubscribe https://github.com/notifications/unsubscribe-auth/AANGW5IJXSQ6GJKW775FB7TU6OAR5ANCNFSM5P73K5BA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
You are receiving this because you are subscribed to this thread.Message ID: @.***>
-- Thanks, Bill
William B. Andreopoulos, Ph.D. Joint Genome Institute LBNL
Thanks for the quick feedback! If you don't have time and accept PRs, I could help, if you'd like
Hello: Apologies for the delay. I just looked at the docker branch code that is used for the docker image, and I am not sure how you would like me to make it more user-friendly. I can think of changing the .sh script name for start. If you would like to add a python script with arg parsing that suits you, I will be glad to accept a PR.
Converting the code into a legit python package would be greatly helpful with install, version management, and usage. Do you have any plans of doing that?
Hi Nick @nick-youngblut : I just released a new GPU docker image billandreo/deeplasmid.tf.gpu2 It is built on tensorflow/tensorflow:latest-gpu and it is much more user-friendly to use: sudo /usr/bin/docker run -it -v /path/to/fasta:/srv/jgi-ml/classifier/dl/in.fasta -v /path/to/OUT/dir:/srv/jgi-ml/classifier/dl/outdir billandreo/deeplasmid.tf.gpu2 deeplasmid.sh in.fasta outdir
I dont have a plan to release as a conda package at the moment, as Docker containers were more my "cup of tea" in the past and I am too busy with other projects and teaching. I added you as a collaborator if you want to make PRs. Thanks
Thanks for creating the docker image!
The benefit of creating a conda package versus docker is the easy integration of deepplasmid into an existing environment (e.g., a conda environment with many python/R packages & bfx tools already installed).
The
Deeplasmid Docker container for GPU
section in the README states to use the following command for running deeplasmid for plasmid identification on GPU:It would be great to have a more user-friendly interface, such as a bash/python script with arg parsing