Skip to main content

A simple utility to execute your deep learning scripts when there are enough idle gpus

Project description

Wait4GPU

A simple utility to execute your deep learning scripts when there are enough idle gpus.

一个在有足够的空闲gpu时执行深度学习训练的小工具

Install

python setup.py install

Usage

As a command wrapper

python3 -m Wait4GPU [--num-required NUM_REQUIRED] [--no-python] training_script

Example:

# train `atss_r50_fpn_1x_coco.py` with 1 gpu using mmdetection
python3 -m Wait4GPU tools/train.py configs/atss/atss_r50_fpn_1x_coco.py

# train `atss_r50_fpn_1x_coco.py` with 2 gpus using mmdetection
python3 -m Wait4GPU --num-required 2 --no-python ./tools/dist_train.sh configs/atss/atss_r50_fpn_1x_coco.py 2

You may refer to Wait4GPU/__main__.py for more detailed usage.

Integrate with your code

from Wait4GPU.wait4gpu import wait_until_idle
import os

available = wait_until_idle(num_req=2)
os.environ["CUDA_VISIBLE_DEVICES"] = ",".join(available)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

wait4gpu-0.0.1.tar.gz (3.7 kB view details)

Uploaded Source

Built Distribution

wait4gpu-0.0.1-py3-none-any.whl (4.4 kB view details)

Uploaded Python 3

File details

Details for the file wait4gpu-0.0.1.tar.gz.

File metadata

  • Download URL: wait4gpu-0.0.1.tar.gz
  • Upload date:
  • Size: 3.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.0.0 CPython/3.11.5

File hashes

Hashes for wait4gpu-0.0.1.tar.gz
Algorithm Hash digest
SHA256 30c129999970d24339befc10bd32144ef73617daa571de3d2baf5bd65b25cfe0
MD5 e1873963cf400917494dcd96091cfdc2
BLAKE2b-256 bbaaa33debefc7223d582d4f4f3694c14dc8e409ecfe597c7e6fee2b9f90d068

See more details on using hashes here.

File details

Details for the file wait4gpu-0.0.1-py3-none-any.whl.

File metadata

  • Download URL: wait4gpu-0.0.1-py3-none-any.whl
  • Upload date:
  • Size: 4.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.0.0 CPython/3.11.5

File hashes

Hashes for wait4gpu-0.0.1-py3-none-any.whl
Algorithm Hash digest
SHA256 154ffdd4c3367db3812f63fda80a070e0d01285e875fe5f6696785e5536a9af2
MD5 e76ac0272e60e9423f3763989347d964
BLAKE2b-256 4028cb0635925cbef66275fca61d0d116f2ae35bb7eb9b3fca8eee93100c47fc

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page