-
Notifications
You must be signed in to change notification settings - Fork 3
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
nextflow-create-docker suggestions #31
Conversation
@@ -98,7 +99,17 @@ In our example here, we will have you build your image using a `requirements.txt | |||
|
|||
Our example will use the files in the [torch_cuda_test directory](https://github.com/uc-cdis/bio-nextflow/tree/master/nextflow_notebooks/containerized_gpu_workflows/torch_cuda_test) of the bio-nextflow repository. You can review the `readme` file in this directory for more information. It is a simple example that will build up from our base image by adding PyTorch. The Nextflow script will ultimately use a python script that checks the version of CUDA in the GPU instance and checks whether it is compatible with the version of PyTorch and CUDA available in the container. | |||
|
|||
First, in the terminal, navigate to the directory where the downloaded Dockerfile and requirements.txt are located. | |||
First, download the contents of the [torch_cuda_test directory](https://github.com/uc-cdis/bio-nextflow/tree/master/nextflow_notebooks/containerized_gpu_workflows/torch_cuda_test). If you have `git`, the easiest way is to clone the repository: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
For this -- I put in the prerequisites that you should have already cloned the repo, so I would rather not put them here. I can add these steps up there, though
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
gotcha, i still think "navigate to the directory where the downloaded Dockerfile and requirements.txt are located" is a bit out of the blue here, the prerequisite are a lot of text ago
maybe "from the directory where you cloned or downloaded the bio-nextflow
repository as a prerequisite, navigate to the torch_cuda_test
directory" and keep the cd
command i added
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I updated this info to reflect your suggestions in a new PR
|
||
[https://quay.io/repository/cdis/containers?tab=tags&tag=latest](https://quay.io/repository/cdis/containers?tab=tags&tag=latest) | ||
[https://quay.io/repository/cdis/containers?tab=tags](https://quay.io/repository/cdis/containers?tab=tags) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I accepted this change but in a new PR
@@ -32,6 +32,7 @@ Some tools you may be using in your workflow can take advantage of GPU capacity | |||
We have 2 images in our current selection that offer [CUDA](https://www.turing.com/kb/understanding-nvidia-cuda) support for running on GPUs -- these have "cuda" in the image name, followed by the CUDA version. When possible, please choose the latest version of CUDA compatible with your tools. | |||
|
|||
> gen3-cuda-12.3-ubuntu22.04-openssl *(preferred)* | |||
> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
i accepted this change, but in a new PR
closing this PR because I incorporated many of these in this PR - #32 |
Improvements