Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
AMBS
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Container registry
Model registry
Operate
Environments
Monitor
Incidents
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
esde
machine-learning
AMBS
Commits
e2b50cd5
Commit
e2b50cd5
authored
3 years ago
by
Michael Langguth
Browse files
Options
Downloads
Patches
Plain Diff
Remove template for running training in container since container-usage is now mandatory.
parent
7f4d9e2f
No related branches found
No related tags found
No related merge requests found
Pipeline
#89873
failed
3 years ago
Stage: build
Stage: test
Stage: deploy
Changes
1
Pipelines
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
video_prediction_tools/HPC_scripts/train_model_era5_container_template.sh
+0
-50
0 additions, 50 deletions
..._tools/HPC_scripts/train_model_era5_container_template.sh
with
0 additions
and
50 deletions
video_prediction_tools/HPC_scripts/train_model_era5_container_template.sh
deleted
100644 → 0
+
0
−
50
View file @
7f4d9e2f
#!/bin/bash -x
#SBATCH --account=deepacf
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --output=train_model_era5_container-out.%j
#SBATCH --error=train_model_era5_container-err.%j
#SBATCH --time=24:00:00
##SBATCH --time=00:20:00
#SBATCH --gres=gpu:1
#SBATCH --partition=booster
#SBATCH --mail-type=ALL
#SBATCH --mail-user=me@somewhere.com
### Two nodes, 8 GPUs
##SBATCH --nodes=2
##SBATCH --ntasks=8
##SBATCH --ntasks-per-node=4
##SBATCH --gres=gpu:4
## Also take care for the job submission with srun below!!!
WORK_DIR
=
`
pwd
`
BASE_DIR
=
$(
dirname
"
$WORK_DIR
"
)
# Name of virtual environment
VIRT_ENV_NAME
=
"my_venv"
# Name of container image (must be available in working directory)
CONTAINER_IMG
=
"
${
WORK_DIR
}
/tensorflow_21.09-tf1-py3.sif"
# clean-up modules to avoid conflicts between host and container settings
module purge
# declare directory-variables which will be modified appropriately during Preprocessing (invoked by mpi_split_data_multi_years.py)
source_dir
=
/my/path/to/tfrecords/files
destination_dir
=
/my/model/output/path
# valid identifiers for model-argument are: convLSTM, savp, mcnet and vae
model
=
convLSTM
datasplit_dict
=
${
destination_dir
}
/data_split.json
model_hparams
=
${
destination_dir
}
/model_hparams.json
# run training in container
export
CUDA_VISIBLE_DEVICES
=
0,1,2,3
## One node, single GPU
srun
--mpi
=
pspmix
--cpu-bind
=
none
\
singularity
exec
--nv
${
CONTAINER_IMG
}
./wrapper_container.sh
${
VIRT_ENV_NAME
}
python3
${
BASE_DIR
}
/main_scripts/main_train_models.py
\
--input_dir
${
source_dir
}
--datasplit_dict
${
datasplit_dict
}
--dataset
era5
--model
${
model
}
--model_hparams_dict
${
model_hparams
}
--output_dir
${
destination_dir
}
/
## Two nodes, 8 GPUs
#srun -N 2 -n 8 --ntasks-per-node 4 singularity exec --nv ${CONTAINER_IMG} ./wrapper_container.sh ${VIRT_ENV_NAME} python3 ${BASE_DIR}/main_scripts/main_train_models.py \
# --input_dir ${source_dir} --datasplit_dict ${datasplit_dict} --dataset era5 --model ${model} --model_hparams_dict ${model_hparams} --output_dir ${destination_dir}/
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment