clearml-docs/docs/release_notes/ver_1_16.md
2024-08-07 10:08:46 +03:00

5.8 KiB

title
Version 1.16

ClearML SDK 1.16.3

New Features and Improvements

  • Add --tags option to clearml-task (ClearML GitHub issue #1284)
  • Add retries parameter to StorageManager.upload_folder() (ClearML GitHub issue #1305)
  • Add clearml-task and CreateAndPopulate support for bash scripts, ipynb, and python modules (requires clearml-agent v1.9+)
  • Add support for HTTP file upload progress reporting
  • Add CLEARML_MULTI_NODE_SINGLE_TASK (values -1, 0, 1, 2) for easier multi-node single Task workloads
  • Add Model.original_task property to models
  • Change Model.task property to return connected task

Bug Fixes

  • Fix Kerastuner framework and examples (ClearML GitHub issue #1279)
  • Fix scalar logging bug with Fire (ClearML GitHub issue #1300)
  • Fix support passing folder to Task.get_script_info() to get the git info
  • Fix Task.launch_multi_node() to enforce the parent of sub-tasks to be the master node 0 task
  • Fix tensorboard numpy 2.0 incompatibility breaks binding
  • Fix Task.launch_multi_node() not supported when used via Pytorch Lightning
  • Fix Jupyter notebook packages and uncommitted changes are sometimes not fetched
  • Fix "can't create new thread at interpreter shutdown" errors (known issue with Python v3.12.0 and other versions)
  • Fix injected task import in Task.populate()
  • Fix dataset with external links will not reuse downloaded data from parents
  • Fix hierarchy for pipeline nodes without args
  • Fix when abort callback is set, set task status to stopped only if running locally, otherwise leave it for the Agent to set it
  • Fix jsonschema / referencing import to include TypeError protection
  • Fix Dataset offline behavior

ClearML Server 1.16.1

Bug Fix

ClearML Server 1.16.0

New Features and Improvements

  • New UI "new experiment creation"
  • ClearML File Server now uses authentication by default. This can be turned off using the fileserver.auth.enabled: false configuration setting
  • Add confirmation pop-up when running experiment is archived

Bug Fixes

  • Fix model plots not displayed in UI model plot tab
  • Fix UI experiment Debug Samples tab freezes when a lot of debug samples are reported in the experiment
  • Fix UI scalar “Export to CSV” button returns an empty CSV file
  • Fix large number of tags in UI tables' tag filter causes UI to crash
  • Fix searching in UI object tables' tag filter causes UI to crash
  • Fix consistent plot colors not maintained when embedded in UI Reports (ClearML GitHub issue #1239)
  • Fix UI plot legends and tooltips wrapping (ClearML Server GitHub issue #241)
  • Fix multiple parameters can be selected for plot axes in scatter plot view in UI experiment comparison
  • Fix image names change in UI task debug sample comparison when a different metric is selected
  • Fix numeric range filters of UI table columns not working
  • Fix queue experiment list is not updated in the UI Orchestration's Queues page
  • Fix My work/Team work persists across UI modules (i.e. projects, datasets, pipelines, etc.)
  • Fix UI single-scalar plots are not ordered alphabetically by scalar name
  • Fix limit on file uploads to the fileserver

ClearML SDK 1.16.2

New Features and Improvements

  • Make dataset preview optional (ClearML GitHub issue #1227)
  • Add api.public_ip_ping (default: 8.8.8.8) and api.public_ip_service_urls (default: ["api.ipify.org";, "ident.me";]) configuration settings to detect public IP and network interface

Bug Fixes

  • Fix python-fire integration (ClearML GitHub issue #1268)
  • Fix path substitution for file:// URIs (ClearML GitHub issues #1217 and #1233)
  • Fix numpy 2.0 compatibility (np.NINF removed)
  • Fix no need to recreate reporter if forking and reporting in subprocess
  • Fix forked detection mechanism

ClearML SDK 1.16.1

Bug Fixes

  • Fix pipeline breaks when continue_on_abort is set to true
  • Fix Pycharm Plugin Windows/Linux interoperability

ClearML SDK 1.16.0

New Features

  • Add additional warning instructing on how to install in case we failed detecting a Jupyter notebook with an import error
  • Add Task.get_executed_queue() to get the name/ID of the queue a task was executed in
  • Move Task.set_resource_monitor_iteration_timeout() to a class method, add wait_for_first_iteration_to_start_sec and max_wait_for_first_iteration_to_start_sec arguments (also add sdk.development.worker.wait_for_first_iteration_to_start_sec and sdk.development.worker.max_wait_for_first_iteration_to_start_sec configuration options)
  • Add support for better pipeline continue behavior including control of children using the continue_behaviour argument
  • Add Python 3.12 support

Bug Fixes

  • Fix pytorch-lightning patches (ClearML GitHub issue #1249)
  • Fix parameter overrides are converted to string when using HPO (ClearML GitHub issue #975)
  • Fix FastAI performance (ClearML GitHub issue #1234)
  • Fix MIG GPU support
  • Fix AMD GPU metrics collection
  • Fix Jupyter password might not be used in some protected JupyterLab instances
  • Fix URL substitution was not applied to registered uploaded files when reporting an event