About horovod

Home: https://github.com/horovod/horovod

Package license: Apache-2.0

Feedstock license: BSD-3-Clause

Summary: Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.

Development: https://github.com/horovod/horovod

Documentation: https://github.com/horovod/horovod

Current build status

Azure

Variant Status linux_64_cuda_compiler_version10.2python3.10. linux_64_cuda_compiler_version10.2python3.8. linux_64_cuda_compiler_version10.2python3.9. linux_64_cuda_compiler_version11.6python3.10. linux_64_cuda_compiler_version11.6python3.8. linux_64_cuda_compiler_version11.6python3.9. linux_64_cuda_compiler_versionNonepython3.10. linux_64_cuda_compiler_versionNonepython3.8. linux_64_cuda_compiler_versionNonepython3.9.

Current release info

Name	Downloads	Version	Platforms

Installing horovod

Installing horovod from the deepmodeling channel can be achieved by adding deepmodeling to your channels with:

conda config --add channels deepmodeling
conda config --set channel_priority strict

Once the deepmodeling channel has been enabled, horovod can be installed with conda:

conda install horovod

or with mamba:

mamba install horovod

It is possible to list all of the versions of horovod available on your platform with conda:

conda search horovod --channel deepmodeling

or with mamba:

mamba search horovod --channel deepmodeling

Alternatively, mamba repoquery may provide more information:

# Search all versions available on your platform:
mamba repoquery search horovod --channel deepmodeling

# List packages depending on `horovod`:
mamba repoquery whoneeds horovod --channel deepmodeling

# List dependencies of `horovod`:
mamba repoquery depends horovod --channel deepmodeling

Updating horovod-feedstock

If you would like to improve the horovod recipe or build a new package version, please fork this repository and submit a PR. Upon submission, your changes will be run on the appropriate platforms to give the reviewer an opportunity to confirm that the changes result in a successful build. Once merged, the recipe will be re-built and uploaded automatically to the deepmodeling channel, whereupon the built conda packages will be available for everybody to install and use from the deepmodeling channel. Note that all branches in the deepmd-kit-recipes/horovod-feedstock are immediately built and any created packages are uploaded, so PRs should be based on branches in forks and branches in the main repository should only be used to build distinct package versions.

In order to produce a uniquely identifiable distribution:

• If the version of a package is not being increased, please add or increase the build/number.
• If the version of a package is being increased, please remember to return the build/number back to 0.

Feedstock Maintainers

• @njzjz