LoginSign Up
← Back to all sparks
C

cloudml

ANALYTICS
Velocity0.0

R interface to Google Cloud Machine Learning Engine

github.com/rstudio/cloudml ↗

Six years since the last functional change, and Google renamed the service it wraps in the release before that

machine-learninggoogle-cloudtensorflowmodel-trainingdormant
◆Current state
cloudml lets R users train keras, tfestimators and tensorflow models on Google's managed machine learning service, tune hyperparameters there, and deploy the results. Its last release with functional content was 0.6.1 in September 2019, which adapted to Google renaming the service from ml-engine to ai-platform. The only entry since is a 2025 documentation update made to satisfy CRAN.
◆Where it's heading
The visible arc is short and stops abruptly. Releases through 2018 tracked the TensorFlow runtime version and patched packaging problems; 0.6.1 added a customCommands hook so users could run OS-level setup before package installation, and adjusted to the service's new name. Then nothing for six years. A 2025 release containing only documentation changes is the standard signal of a package being kept on CRAN rather than being developed.
◆Prediction
There is nothing in this feed to support a prediction of functional work. The most likely next event is another CRAN-driven documentation patch, or archival.

◆Recent moves

  1. 0y ago

    Documentation updated for CRAN

    Documentation updates for CRAN, six years after the previous release. No functional change — this is a package being kept on the archive rather than developed.

    View source ↗
  2. 6y ago

    ai-platform command adopted; custom pre-install commands added

    Switches to the ai-platform command when the installed Google Cloud SDK is recent enough, following the service's rename from ml-engine, and adds a customCommands flag in cloudml.yml for running OS-level setup before packages are installed. The custom-commands hook is the more durable of the two — it is the escape valve for system dependencies the package cannot anticipate.

    View source ↗
  3. 7y ago

    Default runtime moves to TensorFlow 1.9

    Moves the default to the TensorFlow 1.9 runtime. For a package whose job is submitting jobs to a managed service, the default runtime version is the main thing users inherit without asking.

    View source ↗
  4. 8y ago

    Patch for CRAN results and a packrat error

    A patch over 0.5.0 fixing CRAN check results and a missing-packrat error. Packaging repair a week after the previous release.

    View source ↗
  5. 8y ago

    Cloud training, GPU jobs, tuning and deployment from R

    The entry lists the package's capabilities rather than what changed in this version — cloud training for keras, tfestimators and tensorflow models, on-demand GPUs including Tesla P100, hyperparameter tuning, and deployment to Google's prediction service. Read as a statement of scope, it is the fullest description of the package in the feed; read as a changelog, it says nothing about the release.

    View source ↗