Skip to main content

Efficient Large-Scale Distributed Training with Colossal-AI and Lightning AI.

Project description

Lightning ⚡ Colossal-AI

Efficient Large-Scale Distributed Training with Colossal-AI and Lightning AI

lightning PyPI Status PyPI - Python Version PyPI Status Deploy Docs

General checks CI testing Build Status pre-commit.ci status


Installation

pip install -U lightning-colossalai

Usage

Simply set the strategy argument in the Trainer:

import lightning as L

trainer = L.Trainer(strategy="colossalai", precision="16-mixed", devices=...)

For more fine-grained tuning of Colossal-AI's parameters, pass the strategy object to the Trainer:

import lightning as L
from lightning_colossalai import ColossalAIStrategy

strategy = ColossalAIStrategy(...)
trainer = L.Trainer(strategy=strategy, precision="16-mixed", devices=...)

Find all configuration options in the docs!

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

lightning-colossalai-0.1.0.tar.gz (15.9 kB view details)

Uploaded Source

Built Distribution

lightning_colossalai-0.1.0-py3-none-any.whl (14.8 kB view details)

Uploaded Python 3

File details

Details for the file lightning-colossalai-0.1.0.tar.gz.

File metadata

  • Download URL: lightning-colossalai-0.1.0.tar.gz
  • Upload date:
  • Size: 15.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.11.2

File hashes

Hashes for lightning-colossalai-0.1.0.tar.gz
Algorithm Hash digest
SHA256 3e483405012aea6ef8c1d7b9dd04c227dc45ad331f374e6ff3c0ff185f9aaedf
MD5 584d585fc2842b83e5311b2273dd9b84
BLAKE2b-256 2d5b59fdfecd68ad119ec923bac3451956239ca403774541e443eda2a2718377

See more details on using hashes here.

File details

Details for the file lightning_colossalai-0.1.0-py3-none-any.whl.

File metadata

File hashes

Hashes for lightning_colossalai-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 46783c408f52747517a4c28424f03fa3813c5d7accc5aa7d1fed1aac955122a4
MD5 27446c835955d9fe8c2a2e8cb57b957e
BLAKE2b-256 dc148cb492a6ed13a6712d2ccf01aba4250332d0cdf23835299e6394bdfd5b60

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