I have installed at the hydra cluster of the IFT the Planck 2018 likelihoods, CLASS and MontePython.
In MontePython, when I run a script serially then I produce chains as expected. The problem comes when I submit a job to the cluster. Then I get the following error message:
"Missing Library Error: You must have compiled the classy.pyx file. Please go to
/path/to/class/python and run the command python setup.py build"
Following the above steps I have this message:
"running build
running build_ext
skipping 'path/class/python/../python/classy.c' Cython extension (up-to-date)"
but my problem still persists with the message: "Missing library Error ... " when trying to send a job to the cluster. I emphasize that I can produce chains when I run the script serially.
This is my script:
!/bin/sh
SBATCH -J MPtest
SBATCH --partition=batch
SBATCH --get-user-env
SBATCH -o ./job.%j.out
SBATCH -e ./job.%j.err
SBATCH -N 2
SBATCH -n 2
SBATCH -t 3-00:00:00
export OMP_NUM_THREADS=8
prun python montepython/MontePython.py run -p input_RA/base2018TTTEEE.param -o chains_RA -N 15000
Compiler versions:
My gcc version is: 4.8.3
My icc version is: 16.0.0
My python version is: 2.7.13 Entough
How can I solve this issue on sending jobs to the cluster?
Hi,
I have installed at the hydra cluster of the IFT the Planck 2018 likelihoods, CLASS and MontePython.
In MontePython, when I run a script serially then I produce chains as expected. The problem comes when I submit a job to the cluster. Then I get the following error message: "Missing Library Error: You must have compiled the classy.pyx file. Please go to /path/to/class/python and run the command python setup.py build"
Following the above steps I have this message: "running build running build_ext skipping 'path/class/python/../python/classy.c' Cython extension (up-to-date)"
but my problem still persists with the message: "Missing library Error ... " when trying to send a job to the cluster. I emphasize that I can produce chains when I run the script serially.
This is my script:
!/bin/sh
SBATCH -J MPtest
SBATCH --partition=batch
SBATCH --get-user-env
SBATCH -o ./job.%j.out
SBATCH -e ./job.%j.err
SBATCH -N 2
SBATCH -n 2
SBATCH -t 3-00:00:00
export OMP_NUM_THREADS=8
prun python montepython/MontePython.py run -p input_RA/base2018TTTEEE.param -o chains_RA -N 15000
Compiler versions: My gcc version is: 4.8.3 My icc version is: 16.0.0 My python version is: 2.7.13 Entough
How can I solve this issue on sending jobs to the cluster?
Many thanks! Rubén