SharpDove45 you can programmatically control the configured server using https://allegro.ai/clearml/docs/rst/references/clearml_python_ref/task_module/task_task.html?highlight=set_credentials#clearml.task.Task.set_credentials
MysteriousBee56 would providing Trains with an "import mode" (say, via environment or command line variable), which means that it should create a draft server entry, populate all the execution/environment info and exit before it actually starts employing the ML infrastructure address your use case?
HappyDove3 Notice that in https://github.com/allegroai/clearml/issues/400 the goal is to see a table plot in the UI scalars tab for a specific experiment (with additional discussions on how these will be addressed when comparing experiments).
Note that once you take the approach you suggested of logging your metrics single values, you can configure your experiment comparison scalars view to show single values instead of the time-series graph which I think will provide you with the matrix c...
@<1580367723722969088:profile|SmoothDuck83> Not every plot is trivially be formed as a table (i.e. CSV), that's why the JSON export is available for all plots.
What were you considering?
DefeatedCrab47 For the most part, mlflow can serve basic ML models using scikit-learn. In contrast, Trains was designed with more general purpose ML/DL workflows in mind, for which there's no "generic" way to serve models as different scenarios can use different input encoding, models results would be represented in a variety of forms, etc.
Consider also, that creating an HTTP endpoint for model inference is quite a breeze: there are multiple examples of Flask on top of any DL/ML framework w...
Take a look at https://clear.ml/docs/latest/docs/pipelines/pipelines_sdk_tasks#running-the-pipeline ;
By default pipelines are enqueued for execution by a ClearML Agent. You can explicitly change this behaviour in your code.
GentleSwallow91 For more information, look at what ClearML logs for your experiments: https://docs-testing.allegro.ai/docs/latest/docs/fundamentals/task#logging-task-information
ExcitedFish86 You can https://clear.ml/docs/latest/docs/webapp/webapp_exp_table#adding-metrics-and--or-hyperparameters to include any parameter/metric column that helps your analysis (and subsequently filter the table on those columns).
There's not yet the equivalent of a parameter importance visualization, though such insight visualizations are definitely in our sights.
Sure appreciate if you can https://github.com/allegroai/clearml/issues/new on the subject :)
Hi HealthyStarfish45 ,
Since you're discussing the experiment list, I assume that by "fixed view per experiment" you actually mean "per project" (as the list view is across all experiments in the list)?
Under this assumption, note that the view configuration (column sort, custom columns, filters) is also specified in the browser URL. So, until the Trains UI supports in-app per-project view preferences - You can simply bookmark the URL.
Does this help?
UnevenDolphin73 Well... not right now... Currently the ClearML UI only partitions internal artifact types.
That said, having user-defined artifact groups sure sounds worth looking into - Care to https://github.com/allegroai/clearml/issues/new/choose ?
Thanks for clarifying @<1523705301990117376:profile|WickedCat12> .
As I mentioned originally, plotting an arbitrary metric against another is further down the ClearML roadmap.
It'd be great if you use a github issue to help push it through :)
UpsetTurkey67 The single set of online documentation ( https://clear.ml/docs/latest/docs ), denotes OSS/Free-SaaS/Paid features as such. For example: https://clear.ml/docs/latest/docs/configs/clearml_conf#configuration-vault
Hi DefeatedCrab47 ,
The examples folder has just been restructured: Find the example here:
https://github.com/allegroai/trains/blob/master/examples/services/hyper-parameter-optimization/hyper_parameter_optimizer.py
JitteryCoyote63 Great idea. Appreciate if you https://github.com/allegroai/clearml/issues/new/choose .
UnevenDolphin73 I think it'd be easier to track as a separate one.
WittyOwl57 Is that information available for you on each of the compared experiments when you view them individually?
DefeatedCrab47 Thanks for pointing it out.
We'll get in touch with the PyTorch Lightning team to better understand the code restructure they're effecting (see https://github.com/PyTorchLightning/pytorch-lightning/pull/2384 ).
In the mean time, you can look at the prior version: https://github.com/PyTorchLightning/pytorch-lightning/blob/0.8.1/pytorch_lightning/loggers/trains.py
HappyDove3 you can get some more insight on the different configuration methods and how to use theme https://clear.ml/docs/latest/docs/fundamentals/hyperparameters
DepressedChimpanzee34 Have you noticed the "Show n experiments selected" button on the bottom bar? This effectively toggles your view between whatever is currently sorted/filtered and the current item selection.
To address the scenario you describe: Switch to "Show selected experiments", remove the redundant items, and switch back to the original view: "Show all experiments"
Thoughts?
@<1628927672681762816:profile|GreasyKitten62> When you have specific display considerations, you can implement them through report_table's 'extra_layout' and 'extra_data' parameters
@<1559349204206227456:profile|BeefyStarfish55> try checking out the general overview on pipelines here , and info on the pipelines UI here .
Each step's arguments (and results) should appear in the steps details panel (which you could then follow to the underlying task for complete, in-depth, details).
MelancholyElk85 Thanks for calling this to attention. What do you think would have made it easier for you to notice the available extended list content?
I would assume that a "type to match" option would also have helped?
Appreciate if you could https://github.com/allegroai/clearml/issues/new/choose so this can be pushed forward.
RotundHedgehog76 Thanks for the spot - seems like docs are wrong, and CLI help is correct: '--skip-docker-network' will NOT pass '--network host' to the docker.
AverageRabbit65 Adding to SweetBadger76 's reference, e2e examples are available for the different pipeline implementation methods:
https://clear.ml/docs/latest/docs/guides/pipeline/pipeline_controller
https://clear.ml/docs/latest/docs/guides/pipeline/pipeline_decorator
https://clear.ml/docs/latest/docs/guides/pipeline/pipeline_functions
There's an example here to get you going @<1645597514990096384:profile|GrievingFish90> .
We'll definitely look into finding a place for this info in the ClearML docs.
@<1523701157564780544:profile|TenseOstrich47> This is typically indicative of insufficient server disk space causing ES to go into read-only mode or turn active shards into inactive or unassigned (see FAQ ).
The disk watermarks controlling the ES free-disk constraints are defined by default as % of the disk space (so it might look to you like you still have plenty of space, but ES thinks otherwise). You can configure di...
SmarmySeaurchin8 Following up on ColossalDeer61 's hint, notice https://allegroai-trains.slack.com/archives/CTK20V944/p1597248476076700?thread_ts=1597248298.075500&cid=CTK20V944 not-too-old thread on reusing globally installed packages.