deepspeedai / deepspeedai/DeepSpeed
deepspeed.initialize() fails to complete
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
Dear all,
I'm training a custom model on a cluster and have reserved two GPUs which are available (checked it via nvidia-smi).
I'm executing the training script via
SEED=1635 deepspeed --num_gpus=2 train_video.py --config NETWORK_CONFIG.yaml --deepspeed_config Experiment9/ds_config.json
However, deepspeed fails to complete the initialization, i.e. it stays there and is not able to complete the command (there are some prints in the train_video.py so that I can check which commands are completed):
Any idea why this might not work out? Help is very much appreciated!
Thank you so much!
PS: I am quite new to deepspeed, so if any information for answering this question is missing: please let me know -I will provide the missing information immediately.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the reported command, train_video.py, NETWORK_CONFIG.yaml, and Experiment9/ds_config.json; reproduce the initialization hang while collecting the complete output and environment details requested in the issue. Done means identifying a reproducible cause and confirming that deepspeed.initialize() completes across the two GPUs.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- distributed-systems, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100