Reputation
Badges 1
35 × Eureka!Hi ExasperatedCrocodile76 , I guess that you were able to install Scikit-learn and you were able to run it locally, and now you want to try it with an agent on the same machine.
The error is that it can’t find OpenBLAS:
` Run-time dependency openblas found: NO (tried pkgconfig and cmake)
Run-time dependency openblas found: NO (tried pkgconfig)
../../scipy/meson.build:130:0: ERROR: Dependency "OpenBLAS" not found, tried pkgconfig `My question is: did you export some env variabl...
if I were to run an agent that would require to install pandas at some point I’d run it:OPENBLAS="$(brew --prefix openblas)" clearml-agent daemon --queue default
for example I had to do a OPENBLAS="$(brew --prefix openblas)" pip install pandas
to be able to install pandas on my M1 MAC
would the same experiment be called in either clearml server?
right, I’m saying I had to do that in my MAC. In your case you would have to point it to somewhere else. Please check where openblas is installed on your ubuntu
where is the dataset stored? maybe you deleted the credentials by mistake? or maybe you are not installing the libraries needed (for example if using AWS you need boto3, if GCP you need google-cloud-storage)
but it would only be affecting that session in the terminal…so you would want to add it to your .bashrc
in linux you can run in a terminal:export CLEARML_CONFIG_FILE=/new/path/to/conf
just do:import os.path as op dataset_folder = Dataset.get(dataset_id="...").get_local_copy() csv_file = op.join(dataset_folder, 'salary.csv')
so I removed the entrypoint, and now I can see that it tries to install the packages, but it fails because it can’t download the repo
great! and I saw that there were some system packages needed for opencv that were installed automatically that could be turned off. Now I’m just wondering if I could remove the PIP install at the very beginning, so it starts straightaway
would it be possible to change de dataset.add_files to some function that moves your files to a common folder (local or cloud), and then use the last step in the dag to create the dataset using that folder?
Thanks TimelyPenguin76 for your answer! So indeed it was mounting it, and how do I check that “CLEARML_AGENT_SKIP_PYTHON_ENV_INSTALL” is working in my agent in docker?
I’m suggesting MagnificentWorm7 to do that yes, instead of adding the files to a ClearML dataset in each step