Unable to run custom datasets
aflah02 opened this issue · comments
After running all the steps listed to run the code for a custom dataset when I finally ran python style_paraphrase/schedule.py I got the following error -
subprocess.CalledProcessError: Command 'sbatch style_paraphrase/slurm-schedulers/schedule_0.sh' returned non-zero exit status 127.
For the last step I had added "datasets/new_dataset": BASE_CONFIG into DATASET_CONFIG and changed [('dataset',), ["datasets/formality"]] to [('dataset',), ["datasets/new_dataset"]], inside hyperparameters_config.py
I'm not quite sure what the error indicates and how Should I Proceed from here on
Hi @aflah02,
What happens when you run sbatch style_paraphrase/slurm-schedulers/schedule_0.sh on the terminal? Do you have SLURM on your machine?
Oh yes it turns out the server does not have SLURM installed
I'll retry once it's installed
Thanks for the quick reply
Can you kindly tell how to install SLURM? I am working on Colab.
Slurm is optional while using the bash scripts, just run script with bash if you don't have slurm. To install slurm follow https://slurm.schedmd.com/quickstart_admin.html