Skip to main content

Microsoft Azure Batch AI Management Client Library for Python

Project description

Microsoft Azure SDK for Python

This is the Microsoft Azure Batch AI Management Client Library.

Azure Resource Manager (ARM) is the next generation of management APIs that replace the old Azure Service Management (ASM).

This package has been tested with Python 2.7, 3.4, 3.5 and 3.6.

For the older Azure Service Management (ASM) libraries, see azure-servicemanagement-legacy library.

For a more complete set of Azure libraries, see the azure bundle package.

Compatibility

IMPORTANT: If you have an earlier version of the azure package (version < 1.0), you should uninstall it before installing this package.

You can check the version using pip:

pip freeze

If you see azure==0.11.0 (or any version below 1.0), uninstall it first:

pip uninstall azure

Usage

For code examples, see Batch AI Management on docs.microsoft.com.

Provide Feedback

If you encounter any bugs or have suggestions, please file an issue in the Issues section of the project.

Release History

2.0.0 (2018-06-07)

Breaking changes

This version uses 2018-05-01 BatchAI API specification which introduced the following braking changes:

  • Clusters, FileServers must be created under a workspace;

  • Jobs must be created under an experiment;

  • Clusters, FileServers and Jobs do not accept location during creation and belong to the same location as the parent workspace;

  • Clusters, FileServers and Jobs do not support tags;

  • BatchAIManagementClient.usage renamed to BatchAIManagementClient.usages;

  • Job priority changed a type from int to an enum;

  • File.is_directory is replaced with File.file_type;

  • Job.priority and JobCreateParameters.priority is replaced with scheduling_priority;

  • Removed unsupported MountSettings.file_server_type attribute;

  • OutputDirectory.type unsupported attribute removed;

  • OutputDirectory.create_new attributes removed, BatchAI will always create output directories if they not exist;

  • SetupTask.run_elevated attribute removed, the setup task is always executed under root.

Features

  • Added support to workspaces to group Clusters, FileServers and Experiments and remove limit on number of allocated resources;

  • Added support for experiment to group jobs and remove limit on number of jobs;

  • Added support for configuring /dev/shm for jobs which use docker containers;

  • Added first class support for generic MPI jobs;

  • Added first class support for Horovod jobs.

1.0.1 (2018-04-16)

Bugfixes

  • Fix some invalid models in Python 3

  • Compatibility of the sdist with wheel 0.31.0

1.0.0 (2018-03-19)

General Breaking changes

This version uses a next-generation code generator that might introduce breaking changes.

  • Model signatures now use only keyword-argument syntax. All positional arguments must be re-written as keyword-arguments. To keep auto-completion in most cases, models are now generated for Python 2 and Python 3. Python 3 uses the “*” syntax for keyword-only arguments.

  • Enum types now use the “str” mixin (class AzureEnum(str, Enum)) to improve the behavior when unrecognized enum values are encountered. While this is not a breaking change, the distinctions are important, and are documented here: https://docs.python.org/3/library/enum.html#others At a glance:

    • “is” should not be used at all.

    • “format” will return the string value, where “%s” string formatting will return NameOfEnum.stringvalue. Format syntax should be prefered.

  • New Long Running Operation:

    • Return type changes from msrestazure.azure_operation.AzureOperationPoller to msrest.polling.LROPoller. External API is the same.

    • Return type is now always a msrest.polling.LROPoller, regardless of the optional parameters used.

    • The behavior has changed when using raw=True. Instead of returning the initial call result as ClientRawResponse, without polling, now this returns an LROPoller. After polling, the final resource will be returned as a ClientRawResponse.

    • New polling parameter. The default behavior is Polling=True which will poll using ARM algorithm. When Polling=False, the response of the initial call will be returned without polling.

    • polling parameter accepts instances of subclasses of msrest.polling.PollingMethod.

    • add_done_callback will no longer raise if called after polling is finished, but will instead execute the callback right away.

Features

  • added support for job level mounting

  • added support for environment variables with secret values

  • added support for performance counters reporting in Azure Application Insights

  • added support for custom images

  • added support for pyTorch deep learning framework

  • added API for usage and limits reporting

  • added API for listing job files in subdirectories

  • now user can choose caching type during NFS creation

  • get cluster now reports a path segment generated for storing start task output logs

  • get job now reports a path segment generated for job’s output directories

  • renamed EnvironmentSetting to EnvironmentVariable

0.2.0 (2017-10-05)

  • credentials_info property got renamed to credentials.

  • removed unused class FileServerStatus and Code enum

  • renamed enums for CachingType and VmPriority

  • removed ‘statuses’ attribute on FileServer

0.1.0 (2017-10-03)

  • Initial Release

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

azure-mgmt-batchai-2.0.0.zip (185.4 kB view details)

Uploaded Source

Built Distribution

azure_mgmt_batchai-2.0.0-py2.py3-none-any.whl (174.3 kB view details)

Uploaded Python 2 Python 3

File details

Details for the file azure-mgmt-batchai-2.0.0.zip.

File metadata

File hashes

Hashes for azure-mgmt-batchai-2.0.0.zip
Algorithm Hash digest
SHA256 f1870b0f97d5001cdb66208e5a236c9717a0ed18b34dbfdb238a828f3ca2a683
MD5 6235495fa4c83bac95d7c84699b02a3f
BLAKE2b-256 fa7f0a9e5aa22ea91db0771c267c4815396516177702a4a4eea389eed7af47dd

See more details on using hashes here.

File details

Details for the file azure_mgmt_batchai-2.0.0-py2.py3-none-any.whl.

File metadata

File hashes

Hashes for azure_mgmt_batchai-2.0.0-py2.py3-none-any.whl
Algorithm Hash digest
SHA256 b5f7df6a77fde0bd6b486762eb2c81750b6f1730ee1116689d2dfbd3e03dba95
MD5 6b33a615409a5db4051ebdaca97b9a4e
BLAKE2b-256 d9a5ab796c2a490155c14f9ac4240724ca5c56723315d4dc753030712e6f2e80

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page