mirror of
https://github.com/clearml/clearml-docs
synced 2025-01-31 14:37:18 +00:00
5.9 KiB
5.9 KiB
title |
---|
Version 1.16 |
ClearML Server 1.16.2
Bug Fix
- Fix no graphs are shown in Workers and Queues screens
ClearML SDK 1.16.3
New Features and Improvements
- Add
--tags
option toclearml-task
(ClearML GitHub issue #1284) - Add
retries
parameter toStorageManager.upload_folder()
(ClearML GitHub issue #1305) - Add
clearml-task
andCreateAndPopulate
support for bash scripts, ipynb, and python modules (requiresclearml-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 includeTypeError
protection - Fix Dataset offline behavior
ClearML Server 1.16.1
Bug Fix
- Fix user credentials reset on API server restart (ClearML Server GitHub issue #250)
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
) andapi.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 totrue
- 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, addwait_for_first_iteration_to_start_sec
andmax_wait_for_first_iteration_to_start_sec
arguments (also addsdk.development.worker.wait_for_first_iteration_to_start_sec
andsdk.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