Loading checkpoint shards fails. Leaked semaphore objects

[cohencoo]

This issue occurs when running the run_localGPT.py file.

I've tried both cpu and cuda devices, but still results in the same issue below when loading checkpoint shards. The warning itself can be suppressed, but the process still gets killed no matter what.

Would changing the model_id to another model help? Otherwise I cannot figure out what is causing this vague issue.

Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s]zsh: killed /Users/cohen/Downloads/localGPT-main/.venv/bin/python (.venv) cohen@cohens-mbp localGPT-main % /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown warnings.warn('resource_tracker: There appear to be %d '

[alfawal]

Python's memory usage goes ~33 GB and crashes with that error.

[vicendominguez]

same same ...

MacOS Ventura 13.4 (22F66) + Python 3.11.3 + Apple M1 Pro 16 GB Ram

Smelling like something with the size of the model... no idea.

[subinsebastien]

Same problem on my machine too...

Software:

System Software Overview:

  System Version: macOS 13.3.1 (a) (22E772610a)
  Kernel Version: Darwin 22.4.0
  Boot Volume: Macintosh HD
  Boot Mode: Normal
  Secure Virtual Memory: Enabled
  System Integrity Protection: Enabled
  Time since boot: 10 days, 10 hours, 33 minutes

Hardware:

Hardware Overview:

  Model Name: MacBook Air
  Model Identifier: MacBookAir10,1
  Model Number: MGN63HN/A
  Chip: Apple M1
  Total Number of Cores: 8 (4 performance and 4 efficiency)
  Memory: 8 GB
  System Firmware Version: 8422.100.650
  OS Loader Version: 8422.100.650
  Activation Lock Status: Enabled

[penthoy]

GPT/DB Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Killed same here, I'm on 64gb cpu ram and 16gb Vram, crashed only said "killed" I'm on ubuntu WSL

[johnm-starling]

For Mac users, try looking at the post I made in a related ticket and let me know if it helps.

https://github.com/PromtEngineer/localGPT/issues/25#issuecomment-1578921662