[ncclementi]

• Closes Replace references to RAPIDS to NVIDIA CUDA-X for data science #748

This PR was fully done with and agent. I developed a SKILL to be applied and it generates a report on the untouched RAPIDS appearances with the respective reasons. See this report below.

NOTES:

• For the pages that contain cards the skill replaces RAPIDS->CUDA-X and it adds an intro sentence in the page that references NVIDIA CUDA-X for data science libraries, to be more explicit, but keep the cards short.

After the skill runs.

RAPIDS terminology audit (0 actionable, 84 retained, 11 protected)

RAPIDS terminology audit

• Documentation files scanned: 87
• Actionable prose occurrences: 0
• Retained prose exceptions: 84
• Protected non-prose occurrences: 11
• Repository: deployment

Actionable prose

None. The migration has no unresolved standalone RAPIDS prose.

Retained prose exceptions

The following prose remains unchanged because it matches an approved name or artifact exception.

Approved official name (9)

Retained because the wording is an approved official product or tool name.

• source/custom-docker.md:211

  After starting your container, you can quickly verify that the libraries are installed and working. The container launches directly into a bash shell where you can install the RAPIDS CLI command line utility to verify your installation.

• source/custom-docker.md:225

  2. Install RAPIDS CLI

• source/custom-docker.md:227

  Inside the containers, install the RAPIDS CLI:

• source/custom-docker.md:235

  Once RAPIDS CLI is installed, you can use the rapids doctor subcommand to perform health checks.

• source/guides/rapids-docker-with-cuda.md:335

  After starting your container, you can quickly verify that the libraries are installed and working. The container launches directly into a bash shell where you can install the RAPIDS CLI command line utility to verify your installation.

• source/guides/rapids-docker-with-cuda.md:355

  2. Install RAPIDS CLI

• source/guides/rapids-docker-with-cuda.md:357

  Inside the containers, install the RAPIDS CLI:

• source/guides/rapids-docker-with-cuda.md:365

  Once RAPIDS CLI is installed, you can use the rapids doctor subcommand to perform health checks.

• source/platforms/databricks.md:11

  3. Spark users can install RAPIDS Accelerator for Apache Spark 3.x on Databricks

Package, container, or image version/release; verify context (7)

Retained because the version or release identifies a published package, container, or image artifact; the surrounding context was reviewed.

• CONTRIBUTING.md:157

  The RAPIDS versions for things like container images and install instructions are templated into the documentation pages and are stored in source/conf.py.

• source/cloud/gcp/dataproc.md:36

  At the time of writing Dataproc only supports RAPIDS version 23.12 and earlier with CUDA<=11.8 and Ubuntu 18.04.

• source/cloud/gcp/dataproc.md:38

  Please ensure that your setup complies with this compatibility requirement. Using newer RAPIDS versions may result in unexpected behavior or errors.

• source/cloud/nvidia/brev.md:34

  4. Select "Single Container", choose the RAPIDS release, python and CUDA versions, under the "NVIDIA RAPIDS" Container

• source/platforms/colab.md:16

  The installation process on Colab strives to be an "always working" solution, and sometimes will pin RAPIDS versions to ensure compatibility.

• source/platforms/kubeflow.md:11 (2 occurrences)

  The RAPIDS docker images can be used directly in Kubeflow Notebooks with no additional configuration. To find the latest image head to the installation page, as shown in below, and choose a version of RAPIDS to use. Typically we want to choose the container image for the latest release. Verify the Docker image is selected when installing the latest RAPIDS release.

Packaged suite or container contents; verify context (3)

Retained because the phrase identifies a packaged suite or the contents of an unchanged container; the surrounding context was reviewed.

• source/custom-docker.md:11

  For quick setup with pre-built containers that include the full RAPIDS suite, please see the Official Docker installation guide.

• source/custom-docker.md:151

  | Pre-built RAPIDS | Full RAPIDS suite | 12.9 GB |

• source/hpc.md:118

  Conda installs the full RAPIDS suite alongside the CUDA toolkit in a single

Unchanged container UI label (1)

Retained because the wording is a verified container label shown in a user interface.

• source/cloud/nvidia/brev.md:34

  4. Select "Single Container", choose the RAPIDS release, python and CUDA versions, under the "NVIDIA RAPIDS" Container

Unchanged container label (1)

Retained because the wording is a label for an unchanged container artifact.

• source/custom-docker.md:151

  | Pre-built RAPIDS | Full RAPIDS suite | 12.9 GB |

Unchanged container name (47)

Retained because existing RAPIDS container artifact names have not changed.

• source/_includes/install-rapids-with-docker.md:3

  For this example we are going to run the RAPIDS Docker container so we need to know the name of the most recent container.

• source/_includes/menus/aws.md:49

  Launch the RAPIDS container as a Sagemaker notebook.

• source/_includes/menus/gcp.md:19

  Launch the RAPIDS container in Vertex AI managed notebooks.

• source/cloud/aws/sagemaker.md:84

  • Training script should accept hyperparameters as command line arguments. Starting with the base RAPIDS container (pulled from Docker Hub), use a Dockerfile to augment it by copying your training code and set WORKDIR path to the code.

• source/cloud/azure/azure-vm-multi.md:21

  In Python terminal, a cluster can be created using the dask_cloudprovider package. The below example creates a cluster with 2 workers in westus2 with Standard_NC12s_v3 VMs. The VMs should have at least 100GB of disk space in order to accommodate the RAPIDS container image and related dependencies.

• source/cloud/nvidia/brev.md:72

  Leave Install Jupyter on the host turned off. It installs a separate JupyterLab on the instance on port 8888, the same port the container publishes, and the RAPIDS container already serves JupyterLab with the example notebooks.

• source/cloud/nvidia/brev.md:255

  • Please note: Git is not preinstalled in the RAPIDS container, but can be installed into the container when it is running using

• source/custom-docker.md:8

  This guide provides instructions for building custom RAPIDS Docker containers. This approach allows you to select only the libraries you need, which is ideal for creating minimal, customizable images that can be tuned to your requirements.

• source/custom-docker.md:16

  Building a custom RAPIDS container offers several advantages:

• source/custom-docker.md:145

  One of the key benefits of building custom RAPIDS containers is the significant reduction in image size compared to the pre-built RAPIDS images. Here are actual measurements from containers containing only cuDF:

• source/custom-docker.md:157

  One of the benefits of building custom RAPIDS containers is the ability to easily add your own packages to the environment. You can add any combination of the required libraries and other packages to create a fully featured container for your workloads.

• source/examples/rapids-autoscaling-multi-tenant-kubernetes/notebook.ipynb#cell-38:3

  • First there will be a new prepull Pod scheduled on the new node which will start streaming the RAPIDS container image.

• source/examples/rapids-sagemaker-higgs/notebook.ipynb#cell-10:1

  Download latest RAPIDS container from DockerHub

• source/examples/rapids-sagemaker-higgs/notebook.ipynb#cell-10:3 (2 occurrences)

  To build our RAPIDS Docker container compatible with Amazon SageMaker, you’ll start with base RAPIDS container, which the nice people at NVIDIA have already built and pushed to DockerHub.

• source/examples/rapids-sagemaker-higgs/notebook.ipynb#cell-17:4 (2 occurrences)

  the SageMaker-compatible RAPIDS container to the Amazon Elastic Container Registry (Amazon ECR) to store your Amazon SageMaker compatible RAPIDS container and make it available for Amazon SageMaker.

• source/examples/rapids-sagemaker-higgs/notebook.ipynb#cell-24:1

  Testing your Amazon SageMaker compatible RAPIDS container locally

• source/examples/rapids-sagemaker-higgs/notebook.ipynb#cell-29:1

  Congrats, you successfully trained your Random Forest model on the HIGGS dataset using an Amazon SageMaker compatible RAPIDS container. Now you are ready to run experiments on a cluster to try out different hyperparameters and options in parallel.

• source/examples/rapids-sagemaker-hpo/notebook.ipynb#cell-58:3

  The first step to building our Estimator, is to augment a RAPIDS container with our ML Workflow code from above, and push this image to Amazon Elastic Cloud Registry so it is available to SageMaker.

• source/examples/xgboost-azure-mnmg-daskcloudprovider/notebook.ipynb#cell-1:12

  1. Option 1: Using an Azure Marketplace image made available for free from NVIDIA. The RAPIDS container will be subsequently downloaded once the VMs start up.

• source/examples/xgboost-azure-mnmg-daskcloudprovider/notebook.ipynb#cell-1:13

  2. Option 2: Using packer to create a custom VM image to be used in the cluster. This image will include the RAPIDS container, and having the container already inside the image should speed up the process of provisioning the cluster.

• source/examples/xgboost-azure-mnmg-daskcloudprovider/notebook.ipynb#cell-18:1

  In general, if we use a generic image to create a cluster, we would have to wait till the new VMs are provisioned fully with all dependencies. The provisioning step does several things such as set the VM up with required libraries, set up Docker, install the NVIDIA drivers and also pull and decompress the RAPIDS container etc. This usually takes around 10-15 minutes of time depending on the cloud provider. If the user wants to fire up a cluster quickly, setting up a VM from a generic image every time may not be optimal.

• source/examples/xgboost-azure-mnmg-daskcloudprovider/notebook.ipynb#cell-18:3

  Further, as detailed in Option 1, we can also choose to use a custom Marketplace VM from NVIDIA. However, we will still have to download and decompress the RAPIDS container. So the setup time to start the workers and the scheduler would still be around 8-10 minutes.

• source/examples/xgboost-rf-gpu-cpu-benchmark/notebook.ipynb#cell-4:3

  In order to work with RAPIDS container, the entrypoint logic should parse arguments, load, preprocess and split data, build and train a model, score/evaluate the trained model, and emit an output representing the final score for the given hyperparameter setting.

• source/guides/caching-docker-images.md:11

  This is an example manifest to deploy a Daemonset with the RAPIDS container.

• source/guides/colocate-workers.md:74

  • The container image must contain NVIDIA CUDA-X libraries for data science, the official RAPIDS container images are a good choice for this.

• source/guides/index.md:25

  Building RAPIDS Containers from a custom base image

• source/guides/rapids-docker-with-cuda.md:1

  Building RAPIDS containers from a custom base image

• source/guides/rapids-docker-with-cuda.md:112

  Here are complete examples showing how to build a RAPIDS container with CUDA 13.1.1 components on an Ubuntu 24.04 base image:

• source/guides/scheduler-gpu-optimization.md:120

  • The container image must contain NVIDIA CUDA-X libraries for data science, the official RAPIDS container images are a good choice for this.

• source/hpc.md:214

  to environment modules, as the RAPIDS container image ships with CUDA and all

• source/hpc.md:246

  Import the RAPIDS container image as a squashfs file. We recommend

• source/platforms/coiled.md:30

  The simplest way to get up and running with NVIDIA CUDA-X libraries for data science on Coiled is to launch a Jupyter notebook server using the RAPIDS notebook container.

• source/platforms/coiled.md:98

  We can use this to run GPU code on a remote environment using the RAPIDS container. You can set the coiled CLI to keep the VM around for a few minutes after execution is complete just in case you want to run it again and reuse the same hardware.

• source/platforms/coiled.md:124

  To start an interactive Jupyter notebook session with Coiled Notebooks run the RAPIDS notebook container via the notebook service.

• source/platforms/index.md:26

  Launch RAPIDS containers and cluster on Kubernetes with various tools.

• source/platforms/kubeflow.md:43

  New Kubeflow notebook form named rapids with the custom RAPIDS container image, 2 CPU cores, 8GB of RAM, and 1 NVIDIA GPU selected

• source/platforms/kubeflow.md:46

  Then, you can scroll to the bottom of the page and hit launch. You should see it starting up in your list. The RAPIDS container images are packed full of amazing tools so this step can take a little while.

• source/platforms/kubeflow.md:68

  The RAPIDS container also comes with some example notebooks which you can find in /rapids/notebooks. You can make a symbolic link to these from your home directory so you can easily navigate using the file explorer on the left ln -s /rapids/notebooks /home/jovyan/notebooks.

• source/platforms/kubeflow.md:140

  Next, create a Dask cluster using the KubeCluster class. Set the container image to match the one used for your notebook environment and set the number of GPUs to 1. Also tell the RAPIDS container not to start Jupyter by default and run our Dask command instead.

• source/platforms/kubernetes.md:113

  The RAPIDS container comes with the jupyter-server-proxy plugin preinstalled which you can use to access other services running in your notebook via the Jupyter URL. However, by default this is restricted to only proxying services running within your Jupyter Pod. To access other resources like Dask clusters that have been launched in the Kubernetes cluster we need to configure Jupyter to allow this.

• source/platforms/kubernetes.md:278

  alt: Screenshot of the RAPIDS container running Jupyter showing the nvidia-smi command with a GPU listed

• source/platforms/kubernetes.md:296

  Individual users can also install the Dask Helm Chart which provides a Pod running Jupyter alongside a Dask cluster consisting of Pods running the Dask scheduler and worker components. You can customize this helm chart to run the RAPIDS container images as both the notebook server and Dask cluster components so that everything can benefit from GPU acceleration.

• source/platforms/kubernetes.md:308

  If you are using KubeFlow you can integrate NVIDIA CUDA-X libraries for data science right away by using the RAPIDS container images within notebooks and pipelines and by using the Dask Operator to launch GPU accelerated Dask clusters.

• source/tools/index.md:32

  Launch RAPIDS containers and clusters as native Kubernetes resources with the Dask Operator.

• source/tools/kubernetes/dask-operator.md:54

  • The container image must contain NVIDIA CUDA-X libraries for data science, the official RAPIDS container images are a good choice for this.

Unchanged image name (14)

Retained because existing RAPIDS image artifact names have not changed.

• source/cloud/azure/azureml.md:169

  and to create a Dockerfile to build and image that starts from a RAPIDS image and install additional packages needed for the

• source/custom-docker.md:145

  One of the key benefits of building custom RAPIDS containers is the significant reduction in image size compared to the pre-built RAPIDS images. Here are actual measurements from containers containing only cuDF:

• source/examples/rapids-azureml-hpo/notebook.ipynb#cell-18:1

  We'll be using a custom RAPIDS docker image to setup the environment. This is available in rapidsai/base repo on DockerHub.

• source/examples/rapids-sagemaker-hpo/notebook.ipynb#cell-61:1

  Our container can either be built on top of a RAPIDS image from the nightly or stable release as a starting layer.

• source/examples/rapids-sagemaker-hpo/notebook.ipynb#cell-75:1

  The build step will be dominated by the download of the RAPIDS image (base layer). If it's already been downloaded the build will take less than 1 minute.

• source/examples/time-series-forecasting-with-hpo/notebook.ipynb#cell-110:1

  Choose the same RAPIDS image you used for launching the notebook session

• source/examples/xgboost-azure-mnmg-daskcloudprovider/notebook.ipynb#cell-8:9

  We need to pass in the docker argument docker_args = '--shm-size=256m' to allow larger shared memory for successfully running multiple docker containers in the same VM. This is the case when each VM has more than one worker. Even if you don't have such a case, there is no harm in having a larger shared memory. Finally, note that we use the RAPIDS docker image to build the VM and use the dask_cuda.CUDAWorker to run within the VM. This will run the worker docker image with GPU capabilities instead of CPU.

• source/examples/xgboost-gpu-hpo-job-parallel-k8s/notebook.ipynb#cell-8:1

  Choose the same RAPIDS image you used for launching the notebook session

• source/examples/xgboost-gpu-hpo-mnmg-parallel-k8s/notebook.ipynb#cell-11:1

  Choose the same RAPIDS image you used for launching the notebook session

• source/guides/caching-docker-images.md:9

  To guarantee that each node runs a consistent workload, we will deploy a Kubernetes DaemonSet utilizing the RAPIDS image. This DaemonSet will prevent Dask worker Pods created from this image from entering a pending state when tasks are scheduled.

• source/platforms/kubeflow.md:11

  The RAPIDS docker images can be used directly in Kubeflow Notebooks with no additional configuration. To find the latest image head to the installation page, as shown in below, and choose a version of RAPIDS to use. Typically we want to choose the container image for the latest release. Verify the Docker image is selected when installing the latest RAPIDS release.

• source/platforms/kubeflow.md:142

  This can take a similar amount of time to starting up the notebook container as it will also have to pull the RAPIDS docker image.

• source/platforms/kubernetes.md:9

  For single-user interactive sessions you can run the RAPIDS docker image which contains a conda environment with the libraries and Jupyter for interactive use.

• source/tools/kubernetes/dask-helm-chart.md:45

  [jupyter|scheduler|worker].image.* is updated with the RAPIDS "runtime" image from the stable release,

Verified UI or template label (2)

Retained because the wording is a verified user-interface or template label.

• source/platforms/nvidia-ai-workbench.md:45

  Then scroll down and select "RAPIDS with CUDA" from the list of templates.

• source/platforms/nvidia-ai-workbench.md:47

  

Protected non-prose occurrences

These occurrences are outside prose and were intentionally preserved in code, outputs, URLs, destinations, paths, commands, or identifiers.

Fenced code block (7)

Not prose; preserved as code or literal content.

• source/cloud/aws/ec2.md:80

  --description "RAPIDS EC2 security group" \

• source/cloud/azure/azureml.md:200

  description="RAPIDS environment with azureml-mlflow",

• source/custom-docker.md:246

  🧑‍⚕️ Performing REQUIRED health check for RAPIDS

• source/examples/fraud-detection-mlops-pipeline/notebook.ipynb#cell-4:25

  ├── environment.yml Conda environment (RAPIDS + pipeline deps)

• source/guides/rapids-docker-with-cuda.md:376

  🧑‍⚕️ Performing REQUIRED health check for RAPIDS

• source/hpc.md:167

  help([[RAPIDS {{ rapids_version }} - GPU-accelerated data science libraries.]])

• source/hpc.md:169

  whatis("Name: RAPIDS")

Notebook code cell (4)

Not prose; preserved as executable notebook code.

• source/examples/rapids-azureml-hpo/notebook.ipynb#cell-21:8

  description="RAPIDS environment with azureml-mlflow",

• source/examples/rapids-azureml-hpo/notebook.ipynb#cell-29:25

  sweep_job.description = "Run RAPIDS hyperparameter sweep job"

• source/examples/xgboost-azure-mnmg-daskcloudprovider/notebook.ipynb#cell-23:15

  "dept": "RAPIDS-CSP",

• source/examples/xgboost-azure-mnmg-daskcloudprovider/notebook.ipynb#cell-23:16

  "task": "RAPIDS Custom Image deployment",

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB