NVIDIA NeMo-Run issues - Githubissues

NVIDIA / NeMo-Run

A tool to configure, launch and manage your machine learning experiments.

Apache License 2.0

78 stars 20 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Print the content of slurm script in the sbatch command, so that it's visible in the logs

#68 Kipok opened 2 months ago
0
Fix lint and fmt errors

#67 hemildesai closed 2 months ago
0
Adding PatternPackager to allow packaging without git repo

#66 Kipok closed 2 months ago
0
Pre pull images and display status in docker scheduler

#65 hemildesai closed 2 months ago
0
[BUG] Using the shorthand union type operator `|` in entrypoint type signatures causes universal failures on parsing.

#64 skothenhill-nv opened 2 months ago
0
Fix absolute path in cwd in slurm executor

#63 hemildesai closed 2 months ago
0
Support other packagers in Slurm executor

#62 hemildesai closed 2 months ago
1
Cleanup jobs which have already been waited in a sequential run

#61 hemildesai closed 2 months ago
0
The error message when a factory is ill typed is still incorrect

#60 hemildesai opened 2 months ago
0
Dockers are not being killed when running parallel tasks with DockerExecutor

#59 Kipok closed 2 months ago
0
Logs are not being streamed with exp.run(detach=False, tail_logs=True)

#58 Kipok opened 2 months ago
0
Allow running multiple nemo run tasks in parallel with DockerExecutor

#57 Kipok opened 2 months ago
0
Raise cli errors with full stack trace

#56 hemildesai closed 2 months ago
3
Fix faq table of contents

#55 hemildesai closed 2 months ago
0
Fix bugs with applying plugins

#54 hemildesai closed 2 months ago
0
Adding yaml + lazy execution

#53 marcromeyn closed 1 week ago
1
Set non-zero return code on failure

#52 Kipok opened 2 months ago
0
Fix bug with id for Job Group

#51 hemildesai closed 2 months ago
0
Surface exceptions when loading cli entrypoints

#50 hemildesai closed 2 months ago
0
Fix duplicate logs in sequential executors with no dependency support

#49 hemildesai closed 2 months ago
0
Stream docker pull logs or add progress bar for DockerExecutor

#48 Kipok closed 2 months ago
1
Add option to check for uncommitted changes/untracked files in git packager

#47 hemildesai closed 2 months ago
0
Fix docker logs error when container is marked for removal

#46 hemildesai closed 2 months ago
0
Frequently see this error at the end of the local job for DockerExecutor

#45 hemildesai closed 2 months ago
0
Fix sequential task groups and experiment management for docker executor

#44 hemildesai closed 2 months ago
0
Ability to customize the log location between subtasks inside of a task group.

#43 hemildesai closed 2 weeks ago
0
Allow dependency type to be configured in Slurm executor

#42 hemildesai closed 2 months ago
0
Allow custom job details like log folder and job name

#41 hemildesai closed 2 months ago
1
Have an option to throw an error if there are uncommitted changes when packaging the code

#40 Kipok closed 2 months ago
1
Only ask for confirmation once

#39 marcromeyn closed 2 months ago
0
Check for invalid characters in task/experiment names and raise a clear error

#38 Kipok opened 2 months ago
0
Make an option to run task groups sequentially instead of in-parallel

#37 Kipok opened 2 months ago
1
Make job_name customizable or add ability to reuse code across tasks with different name

#36 Kipok opened 2 months ago
0
Error when launching sequential jobs with task groups

#35 Kipok closed 2 months ago
0
Allow to use afterany in slurm dependencies

#34 Kipok closed 2 months ago
0
Make logs multi-threaded

#33 hemildesai closed 2 months ago
0
Experiment.from_title results in error

#32 Kipok closed 2 months ago
1
Add docker executor support

#31 hemildesai closed 2 months ago
3
[DO NOT MERGE] Create temporary files for doc review

#30 hemildesai opened 2 months ago
0
Add examples to test workflow and fix hello-world notebooks

#29 hemildesai closed 2 months ago
0
Adding io-registration more explicitly

#28 marcromeyn opened 2 months ago
0
e2e perf cli

#27 malay-nagda closed 2 months ago
0
Add support for local execution with docker container and task groups

#26 Kipok closed 2 months ago
1
Fix pitfalls and type annotation errors in README docs

#25 malcolmgreaves closed 3 months ago
1
Add support for group slurm jobs and optimize packaging

#24 hemildesai closed 3 months ago
4
Fix yaml serialization for type(object)

#23 hemildesai closed 3 months ago
0
Allow custom JobPaths in SlurmExecutor

#22 hemildesai closed 3 months ago
3
Allow factory functions to be passed as a dotted-import

#21 marcromeyn closed 3 months ago
0
Start using nemo for CLI entrypoint

#20 marcromeyn closed 2 months ago
1
Use constant workdir /nemo_run in SlurmExecutor

#19 hemildesai closed 3 months ago
0

Previous Next