Comments (6)
@jbweston implemented this 7 months ago. I have never used it myself. From looking at the source-code I have no clue as to why this is happening for you.
from adaptive-scheduler.
@aeantipov, I tried this:
scheduler_kwargs = dict(
num_threads = 4,
extra_scheduler=[
"--exclusive",
"--partition=partition"
],
extra_script="umask 0000", # or ["umask 0000"]
executor_type="process-pool",
)
import adaptive_scheduler
sch = adaptive_scheduler.scheduler.SLURM(cores=1, **scheduler_kwargs)
print(sch.job_script())
which prints:
#!/bin/bash
#SBATCH --ntasks 1
#SBATCH --no-requeue
#SBATCH --exclusive
#SBATCH --partition=partition
export MKL_NUM_THREADS=4
export OPENBLAS_NUM_THREADS=4
export OMP_NUM_THREADS=4
export NUMEXPR_NUM_THREADS=4
umask 0000
/Users/basnijholt/miniconda3/bin/python run_learner.py --n 1 --log-fname ${NAME}-${SLURM_JOB_ID}.log --job-id ${SLURM_JOB_ID} --name ${NAME}
So I think you might be using an old version.
from adaptive-scheduler.
It seems, it's a further dependency that's causing the issue.
from adaptive-scheduler.
I have noticed that this actually still doesn't work.
import concurrent.futures
ex = concurrent.futures.ProcessPoolExecutor()
# Ignores extra_script
run_manager.ioloop.run_in_executor(ex, scheduler.write_job_script, "test")
# Correctly uses extra_script
scheduler.write_job_script("test2")
It probably has to do with the serialization of the scheduler.
from adaptive-scheduler.
Right:
import pickle
print(scheduler.extra_script) # prints correctly
print(pickle.loads(pickle.dumps(scheduler)).extra_script) # prints empty string
from adaptive-scheduler.
Fixed by e7eb7df and I released v0.9.16.
from adaptive-scheduler.
Related Issues (20)
- log explorer looked in `old_logs/logs/logfile.out` instead of `old_logs/logfile.out`
- add helper functions to set SLURM options
- Unicode error on template writing HOT 1
- sorting is broken in the logging widget HOT 1
- forced --no-requeue HOT 1
- Early requeue of jobs HOT 1
- UnicodeDecodeError in _make_default_run_script HOT 1
- Add sequential 'strategy' for BalancingLearner
- Make show failed jobs the default in the log viewer
- add pathos backend
- Ideas 💡
- incompatible with python 3.8 HOT 2
- Add average node starting time to widget 💡 HOT 1
- Allow to only save only dataframes instead of both fnames and derived fnames
- save_dataframe does not write atomically
- Periodically call arbitrary callback HOT 4
- Flood of RuntimeErrors with latest adaptive-scheduler HOT 10
- Calls to scheduler are not asynchronous HOT 1
- # of points is off HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from adaptive-scheduler.