Hmm, so this is kind of a hack for ClearML AWS autoscaling ?
and every instance is running an agent? or a single Task?

MelancholyElk85

After I set base docker for pipeline controller task, I cannot clone the repo...

What do you mean by that?
Also, how do you set the PipelineController base_docker_image (I'm assuming the is needed to run the pipeline logic?!, is that correct?)

PompousBeetle71 just making sure, and changing the name solved it?

@<1558986821491232768:profile|FunnyAlligator17> can you provide the full log ?

Sorry I need the full log ... feel free to DM it to me

You mean Task.clone() API call failed ?

the SDK is unable to see each of the nodes?

Exactly ! I mean I love the idea of "nested" component, but implementation wise this is not trivial, it will also hurt the ability of caching individual component. The workaround is to have all the "business logic" in the pipeline function itself, routing data between components is basically "free". The data does not actually go through the pipeline logic, it only passes reference (unless the pipeline logic actually tries to access the data o...

Hi PanickyMoth78
Hmm yes, I think the StorageManager (i.e. the google storage pythonclinet) also needs a json file with the credentials.
Let me check something

Yes, I mean trains-agent. Actually I am using 0.15.2rc0. But, I am using local files, I mean I clone trains and trains-agent repos and install them. Their versions are 0.15.2rc0

I see, that's why we get the git ref, not package version.

StaleButterfly40 are you sure you are getting the correct image on your TB (toy255) ?

Hover near the edge of the plot, the you should get a "bar" you can click on to resize

VivaciousPenguin66 I have the feeling it is the first space in the URI that breaks the credentials lookup.
Let's test it:
from clearml import StorageManager uri = ' ` Birds%2FTraining/TRAIN [Network%3A resnet34, Library%3A torchvision] Ignite Train PyTorch CNN on CUB200.8611ada5be6f4bb6ba09cf730ecd2253/models/cub200_resnet34_ignite_best_model_0.pt'

original

StoargeManager.get_local_copy(uri)

qouted

StoargeManager.get_local_copy(uri.replace(' ', '%20')) `

You can set torch to be installed last:
post_packages: ["horovod", "torch"]
Which will make sure the "trains-agent" version (the one you specified in the "installed packages" will be installed last.

Hey JitteryCoyote63 I think I need to better explain the config feature:
agent.package_manager.post_packages = ["PyJWT"]Basically this means that IF you have pyjwt in the installation package it will be installed after everything else is installed.
This doesn't mean it will always be installed.
Think for example "horovod" has to be installed after you have TF / PyTorch installed.
(The same goes for "pre_package" and Cython)

QuaintJellyfish58 Notice it tries to access AWS not your minio
"This seems like a bug?! can you quickly verify with previous version ?
Also notice you have to provide the minio section in the clearml.conf so it knows how to access the endpoint:
https://github.com/allegroai/clearml/blob/bd53d44d71fb85435f6ce8677fbfe74bf81f7b18/docs/clearml.conf#L113

To automate the process, we could use a pipeline, but first we need to understand the manual workflow

You need to use tf.summary.image and not summary_ops_v2.image

Fixed on main branch (see github issue), RC later today

Image needs to be in range [0, 1] and not [0, 255] (matplotlib and tensorboard can handle either one)

Is there a code to reproduce ?

Hi OddAlligator72

for instance - remove all the metrics from some step onward? 

(I think that as long as the Task is not published you could do such a thing directly with the RestAPI (aka APIClient from python)
What's the use case?

JitteryCoyote63 did you use the new implementation:
https://github.com/allegroai/trains/blob/master/trains/automation/aws_auto_scaler.py

JitteryCoyote63 I think this only holds for the conda distribution.
(Actually quite interesting, I wonder what happens if you already installed cudatoolkit...)

🤞

Hi VexedCat68

txt file or pkl file?

If this is a string , it just stored it (not as a file, this is considered a "link")
https://github.com/allegroai/clearml/blob/12fa7c92aaf8770d770c8ed05094e924b9099c16/clearml/binding/artifacts.py#L521

Hi RipeGoose2
So the http://app.community.clear.ml already contains it.
Next release of the standalone server (a.k.a clearml-server) will include it as well.
I think the ETA is end of the year (i.e. 2 weeks), but I'm not sure on the exact timeframe.
Sounds good ?

you can run md5 on the file as stored in the remote storage (nfs or s3)

s3 is implementation specific (i.e. minio weka wassaby etc, might not support it) and I'm actually not sure regrading nfs (I mean you can run it, but it actually means you are reading the data, that said, nfs by definition I'm assuming is relatively fast access)
wdyt?

Yeah the doctring is always the most updated 🙂

TrickySheep9
you are absolutely correct 🙂

Hi OddShrimp85
If you pass 'output_uri=True' to task init, it will upload the model automatically, or as you said manually with outputmodel class

This is a Sagemaker notebook instances

Yes I think this is the issue

Hi  SmoggyGoat53
There is a storage limit on the file server (basically 2GB per file limit), this  is the cause of the error.
You can upload the 10GB to any S3 alike solution (or a shared folder). Just set the "output_uri" on the Task (either at Task.init or with Task.output_uri = " s3://bucket ")

It uses only one CPU core, could I use multiprocessing somehow?

Hi EcstaticMouse10
Hmm, yes it should be multi core:
https://github.com/allegroai/clearml/blob/a9774c3842ea526d222044092172980ae505e24f/clearml/datasets/dataset.py#L1175
wdyt?